The distribution, characteristics and behaviour of mass movements
triggered by the Kashmir Earthquake 2005, NW Himalaya,
Pakistan
Die Verbreitung, Eigenschaften und das Verhalten von
Massenbewegungen, die durch das Erdbeben in Kaschmir 2005
ausgelöst wurden, NW Himalaya, Pakistan
Der Naturwissenschaftlichen Fakultät
der Friedrich-Alexander-Universität Erlangen-Nürnberg
zur
Erlangung des Doktorgrades Dr. rer. nat.
vorgelegt von
Muhammad Basharat
aus
Muzaffarabad, Pakistan
2012
ii
Als Dissertation genehmigt von der Naturwissenschaftlichen Fakultät
der Friedrich-Alexander-Universität Erlangen-Nürnberg
Tag der mündlichen Prüfung : 10.05.2012 Vorsitzender der:
Promotionskommission: Prof. Dr. Rainer Fink
Erstberichterstatter: Prof. Dr. Joachim Rohn
Zweitberichterstatter: Prof. Dr. Michael Moser
iii
Dedicated to sweet memories of my beloved father (Late)
iv
Acknowledgements
First of all my special gratitude is to my supervisor Professor Dr. Joachim Rohn for giving me an
opportunity to work under his supervision. His valuable guidance and suggestions throughout the research
work kept my confidence and spirit high. Throughout this work he provided me a good environment,
special care, many good ideas, valuable comments, and critical review of this thesis.
I owe my deepest gratitude to Professor Dr. Michael Moser for his kind help, valuable suggestions,
guidance, and critical review of this thesis.
I am grateful to Professor Dr. Mirza Shahid Baig, Institute of Geology, University of Azad Jammu and
Kashmir for his co-supervision during the field work.
I sincerely thank to Dr. Dominik Ehret, who facilitating me to have a good start of my PhD and for his
guidance during the field.
I greatly acknowledged the University of Azad Jammu and Kashmir, Muzaffarabad and Higher Education
Commission (HEC) Pakistan, for granting me a scholarship under Faculty Development Program for my
PhD research work.
I would like to acknowledge Bundesanstalt für Geowissenschaften und Rohstoffe (BGR) and Geological
Survey of Pakistan (GSP) for providing me the satellite data used in this study. I also acknowledged
Planning and Development department AJK, Muzaffarabad for providing the topographical maps.
I sincerely express my gratitude to Professor Dr. Sabir Khan, director Institute of Geology University of
Azad Jammu and Kashmir for providing transport facility during the field work. I extend my thanks to
Professor Dr. Rustam Khan for his kind support and help. I also want to thanks all my colleagues of the
Institute of Geology, University of Azad Jammu and Kashmir.
I greatly appreciate the various kind of assistance from Mrs. Beate Wuttke during my stay at GeoZentrum
Nordbayern. I extend my thanks to Mrs. Gudrun Klein for providing the literature during the research work.
I am grateful to my colleagues Luo Jin, Markus Schleier (special thanks for the German translation of the
abstract), Bi Renneng, Johannes Wiedenmann, Jiang Jiwei and Zeng Bin for their cooperation and
providing me a good environment during my work. I am also thankful to my friend Mr. Basharat Ahmed
for his valuable suggestions and cooperation during the work.
Thanks also to my friends Dr. Nasir Khan, Dr. Aziz, and Mr. Adeel Nasir providing good company during
my stay in Erlangen.
I would like to express deep gratitude to my father (who did not live to see the end of this work) for his
love, affection and support throughout my life. With all my heart I thanks to my all family members,
especially my elder brother Khawaja Muhammad Nasim Advocate, who kept in touch with me during the
whole period and took care of my family during my absence.
Last, but not least my heartfelt love to my wife Zahida and my children Aimen, Wasif and Ayan for their
patience and understanding about my work and spent four years without my love and care.
Muhammad Basharat
v
Kurzfassung
Die vorliegende Arbeit beschäftigt sich mit der Verteilung, den Eigenschaften und dem Verhalten
von Massenbewegungen, die bei dem Erdbeben in Kaschmir 2005 auslöst wurden. Das
Hauptaugenmerk der Arbeit liegt darauf, die statistischen und empirischen Beziehungen zwischen
der Massenbewegungsverteilung und den geologischen, sowie geometrischen Kenndaten zu
analysieren. Zusätzlich wurden Bestandsverzeichnisse von Massenbewegungs Muzaffarabad-
Stadt, sowie den Gebieten um das Jhelum-Tal und das Neelum-Tal angefertigt und in Karten
dargestellt. Zu diesem Zweck wurden Geländeuntersuchungen mit der Interpretation von
Satellitendaten kombiniert. Insgesamt 1.460 Massenbewegungsereignisse wurden aus den SPOT-
Satellitenbildern identifiziert und interpretiert. Von diesen wurden 127 während der
Geländeuntersuchungen vor Ort dokumentiert.
Weiterhin wurde bei diesen eine Klassifizierung in fünf Typen von Massenbwegungen „rock
avalanches, rock falls, debris falls, slump and landslides“ nach Varnes (1978) vorgenommen.
Die Bestandsverzeichnissse zeigen an, dass 3,9% des untersuchten Gebietes von Erdbeben
induzierten Massenbewegungen betroffen ist.
Ein Kapitel dieser Arbeit erörtert vier großmaßstäbliche Massenbewegungen als Fallbeispiele, die
während des Erdbebens aktiviert oder reaktiviert wurden. Diese Ereignisse wurden bezüglich der
geologischen, strukturellen, geometrischen und geotechnischen Charakteristika analysiert. Im
Rahmen des Fallbeispiels des Bergsturzes Hattian Bala („Sturzstrom“) wurde die
Massenbewegung, basierend auf detaillierten Geländeuntersuchungen, geologisch und strukturell
im Maßstab 1:10.000 kartiert. Die Auswertung zeigt, dass die Geometrie und die Versagensart
dieser reaktivierten Massenbewegung sehr stark von der Tektonik und der Lithologie gesteuert
wurden.
Die anderen drei bedeutenden Massenbewegungen, wie der Langarpura und der Neelidandi
Bergsturz, sowie die Massenbewegung Panjgran wurden zum ersten Mal in dieser Studie
ausführlich untersucht und dokumentiert. Die Reaktivierung dieser Massenbewegungen am
hangenden Block der reaktivierten Muzaffarabad-Störung und nahe des Epizentrums legt nahe,
dass die Bodenbewegung, die diese Massenbewegungen ausgelöst hat, hier besonders hoch war.
Um den Zusammenhang zwischen der Verteilung der Massenbewegungen und ihren ursächlichen
Faktoren, wie etwa Abstand zum Epizentrum und Abstand zu aktivierten Störungen,
topographische Parameter (Hangneigung, Exposition, Höhe) und geologische Einheiten zu
untersuchen, wurde eine statistische Analyse vorgenommen. Die Karte der
Massenbewegungsverteilung wurde anhand von SPOT Satellitenbildern unter Einbezug von
vi
eigenen Geländeaufnahmen erstellt. Die Entfernung der Massenbewegungen zum Epizentrum und
zur Entfernung aktivierter Störungen wurde berechnet. Der Index für die Massenbewegungs-
Konzentration bei den statistischen Untersuchungen wurde anhand der Anzahl von
Massenbewegungen pro km² berechnet. Die Berechnung der topographischen Parameter, wie
Hangneigung, Exposition und Höhe erfolgte unter Verwendung eines ASTER basierten digitalen
Höhenmodells (DEM). Die geologischen Parameter wurden bezüglich der lithologischen
Charakteristika untersucht, um den Einfluss der Lithologie auf die Massenbewegungs-
Konzentration zu verstehen.
Die Massenbewegungs-Konzentration ist in der kambrischen Muzaffarabad-Formation aufgrund
des Auftretens von stark zerrütteten, zerscherten Dolomiten und des Auftretens im Nahbereich des
Hangenden Blocks der aktiven Muzaffarabad-Störung besonders hoch.
Die Analyse führt zu der Schlussfolgerung, dass die Verteilung der durch das Erdbeben
ausgelösten Massenbewegungen viel mehr von der Entfernung zur Erdbebenquelle (Epizentrum,
Störung) abhängig ist, als von den topographischen Parametern und den geologischen Einheiten.
Eine empirische Analyse bezüglich verschiedener Parameter von Felsstürzen wurde anhand von
103 Massenbewegungsereignissen ausgeführt. Der Zusammenhang zwischen wichtigen
Parametern, wie etwa Volumen, Fahrböschungswinkel, Schattenwinkel, Hangneigung der
Felssturzhalde, Fallhöhe, Oberflächengrösse und der Reichweite wurde analysiert. Die Analyse
liefert zum einen ein besseres Verständnis betreffend der physikalischen Charakteristika und dem
Verhalten der Massenbewegungen, die durch das Kashmir Erdbeben von 2005 ausgelöst wurden,
und stellt darüber hinaus bedeutende Datensätze zur Gefahren- und Risikobeurteilung bereit. Die
Analyse zeigte, dass der Fahrböschungswinkel teilweise vom Volumen der Massenbewegung
abhängig ist. Allerdings beeinflussen auch die anderen Faktoren, wie Hangsteilheit, hohes Relief
des Gebietes sowie die geologischen Gegebenheiten die Reichweite der untersuchten
Massenbewegungen. Darüber hinaus sind die Werte für den Fahrböschungswinkel bei einigen
Massenbewegungen mit kleinem Volumen, verglichen mit einigen größeren Bewegungen, sehr
hoch. Die Analyse zeigt einen linearen Trend zwischen Fallhöhe und Reichweite.
Zwischen dem Fahrböschungswinkel und Fallhöhe sowie der Reichweite wurden keine
Zusammenhänge gefunden. Das deutet darauf hin, dass der Fahrböschungswinkel keinen direkten
Einfluss auf die Reichweite und die Fallhöhe hat. Anhand von Schattenwinkel und Hangneigung
der Schutthalde zeigen die Analysen keinen klaren Zusammenhang zwischen dem Volumen der
Massenbewegung und der Reichweite.
Beim Vergleich von hier vorgestellten Daten des Kashmir Erdbebens stimmen die Ergebnisse
größtenteils mit publizierten Daten aus anderen Teilen der Erde überein.
vii
Abstract
The presented study deals with the distribution, characteristics and behaviour of mass movements
triggered by the Kashmir earthquake 2005, in Kashmir region of northern Pakistan. The main
focus of this work was to analyze the statistical and empirical relationships between mass
movement distribution and geological and geometrical features. In addition, mass movement
inventory maps of Muzaffarabad city and the areas around the Jhelum valley and the Neelum
valley were produced. For this purpose, ground based field investigations were combined with
satellite data. In total 1,460 mass movement events were identified and interpreted from SPOT
imageries. Of these, 127 mass movements were documented during the ground based investigation
and classified into five types such as “rock avalanches, rock falls, debris falls, slump and
landslides” according to Varnes (1978). The inventory maps indicate that 3.9 % of the area was
affected by earthquake induced mass movements. The analysis suggests that the diverse
geological, tectonic, seismic, geotechnical and morphological conditions control the extent and
geographical distribution of the mass movements.
A section of this study is discussing four large scale mass movements activated or reactivated
during the earthquake, as case studies. These events were analyzed in terms of geological,
structural, geometrical and geotechnical characteristics. In the case study of the Hattian Bala rock
avalanche, the mass movement was mapped geologically and structurally at a scale of 1:10,000
based on detailed field investigations. The analysis indicates that the geometry and failure mode of
this reactivated mass movement was strongly controlled by tectonics and lithology.
The other three significant mass movements such as the Langarpura and Neelidandi rock falls and
the Panjgran slump and rock fall were investigated and documented circumstantial for the first
time in this study. The reactivation of these mass movements on the hanging wall block of the
reactivated Muzaffarabad Fault and close to the epicentral area suggests that the ground motion
triggering these mass movements was very high.
The statistical analysis was performed to analyze the relationship between the mass movement
distribution and their causal factors such as source of earthquake (epicenter, Muzaffarabad Fault),
topographic parameters (slope, aspect, elevation) and geological units. The mass movement
distribution map was prepared using SPOT images incorporated with field data. The mass
movement distances were calculated by using two distance definitions (distance to epicenter and
distance to Muzaffarabad Fault). The index of mass movement concentration for statistical
analysis was calculated based on the number of mass movements per km2. Topographical
parameters such as slope angle, slope aspect and elevation were calculated by using a ASTER
based DEM. The geological parameters were examined according to their lithological
viii
characteristics to understand the influence of lithology on the mass movement concentration. The
mass movement concentration was notably high in the Cambrian Muzaffarabad Formation due to
the occurrence of highly fractured, sheared dolomites situated on the hanging wall block of the
active Muzaffarabad Fault.
The analysis leads to the conclusion that the distribution of mass movements triggered by the
earthquake is mainly depending on the distance from the earthquake source (epicenter,
Muzaffarabad Fault) rather than on topographical parameters and geological units.
An empirical analysis among various parameters of rock falls was conducted for 103 mass
movement events. The relationship among important parameters such as volume, Fahrböschung
angle, shadow angle, talus slope angle, height of fall, travel distance, surface area on the mass
movements travel distance were analyzed. The analyses gives a better understanding regarding
physical characteristics and behaviour of the mass movements triggered by the Kashmir
earthquake 2005, and provides a significant set of data for the hazard and risk evaluation. The
analysis showed that the Fahrböschung angle depends to some extent on the volume of the mass
movements. However, the other factors such as slope steepness, high relief of the area and
geological conditions affect also the travel distance of the investigated mass movements.
Moreover, the Fahrböschung angle values are very high for some mass movements with small
volumes compared to some larger ones. The analysis indicates that there is a linear trend between
the height of fall and travel distance.
No relationships were found between the Fahrböschung angle values and the height of fall, and the
travel distances. This indicates that Fahrböschung angle has no direct influence on the travel
distance and the height of fall. The analyses do not show any clear relationship between the
volume of the mass movements and the travel distance by means of shadow angle and talus slope
angle. The comparisons of Kashmir earthquake data with published data of other parts of the world
are mainly consistent with each other.
List of Contents
ix
CONTENTS 1. Introduction.................…………………………………………………………………….1
1.1. Backdrop……………………………………………………………………………….1
1.2. Literature review………………………………………………………………………..3
1.3. Research area…………………………………………………………………………...4
1.4. Aim of the study………………………………………………………………………..6
1.5. Thesis structure………………………………………………………………................7
2. Regional tectonics..……………………………………………………………………........9
2.1. Tectonics of Himalayas………………………………………………………………...9
2.2. Tectonic setting of the NW Himalayas of Pakistan…………………………..............11
2.3. Structural setting.……………………………………………………………………..14
2.3.1. Main Karakorum Thrust (MKT)……………………………………………….14
2.3.2. Main Mantle Thrust (MMT)…………………………………………...............14
2.3.3. Main Central Thrust (MCT)……………………………………………………14
2.3.4. Panjal Thrust (PT)……………………………………………………...............14
2.3.5. Main Boundary Thrust (MBT)….……………………………………………...15
2.4. Indus Kohistan Seismic Zone (IKSZ)………………………………………................16
2.5. Hazara Kashmir Syntaxis (HKS)……………………………………………...............17
2.6. Tectonic model for the origin of HKS……………………...........................................17
2.7. Core structures of the HKS…………………………………........................................18
2.7.1. Muzaffarabad Anticline….……………………………………………..............19
2.7.2. Muzaffarabad Fault (MF)….…………………………………………………...20
2.7.3. Jhelum Fault (JF)…….………………………………………………………....21
3. Geological setting..………………………………………………………………………...23
3.1. General description……………………………………………………………………23
3.2. Geology of Hazara Kashmir Syntaxis………………………………………………...23
3.3. Geological setting of the study area…………………………………………………..25
3.3.1. Hazara Formation….…………………………………………………………...26
3.3.2. Tanol Formation….………………………………………………………….....28
3.3.3. Muzaffarabad Formation….…………………………………………................28
3.3.4. Manshera Granite………………………………………………………………29
3.3.5. Panjal Formation....………………………………………………….................29
3.3.6. Paleocene-Eocene sequence….………………………………………………...30
3.3.6.1. Hangu Formation…..………………………………………………….30
List of Contents
x
3.3.6.2. Lockhart Formation…..…………………………………………….....30
3.3.6.3. Patala Formation…..………………………………………..................31
3.3.6.4. Margala Hill Formation…..…………………………………………...31
3.3.6.5. Chorgali Formation….………………………………………………..32
3.3.6.6. Kuldana Formation…..………………………………………………..32
3.3.7. Murree Formation...………………………………………………...................33
3.3.8. Kamlial Formation...……………………………………..................................34
3.3.9. Quaternary sediments...……………………………………………………….34
3.3.9.1. Alluvium deposit………………………………………….................34
3.3.9.2. Colluvium deposit……………………………………….. ……….....35
4. Methodology.………………………………………………………………………………36
4.1. Introduction …………………………………………………………………………...36
4.2. Available resources.……………………………………………………………………36
4.2.1. Literature collection.…………………………………………………………….36
4.2.2. Topographic map.…………………………………………………………….…36
4.2.3. Geological map.…………………………………………………………………38
4.2.4. Satellite imagery..………………………………………………….....................38
4.2.5. Digital Elevation Models (DEMs)…...................................................................38
4.3. Field survey.………………………………………………………………...................38
4.4. Database inventory..…………………………………………………………………...43
4.5. Data analysis.…………………………………………………………………………..44
5. Mass movements triggered by the Kashmir earthquake 2005.…………………….......46
5.1. General overview.……………………………………………………………………...46
5.1.1. Damages caused by mass movements….………………………………………46
5.1.2. Earthquake induced mass movements.…………………………………………47
5.2. Terminology and classification of mass movements.…………………………….........48
5.2.1. Classification system of Varnes (1978)..………………………………………..49
5.2.2. Classification system of Cruden and Varnes (1996)..…………………………..49
5.2.3. Classification system used for this study….……………………………………50
5.3. Types of mass movements induced by the earthquake.………………………………..50
5.3.1. Shallow mass movements on very steep slopes.…..…………………………….51
5.3.2. Deep seated mass movements.…..………………………………………………52
5.4. Mass movement identification and classification..……………………………………..53
5.4.1. Landslides..….…………………………………………………………………...53
List of Contents
xi
5.4.1.1. Rotational landslides…………………………………………………54
5.4.1.2. Translational landslides……………………………………………...55
5.4.1.3. Occurrence of landslides in the study area…………………………..55
5.4.2. Rock falls.……………………………………………………………………56
5.4.2.1. Occurrence of rock falls in the study area...………………………...56
5.4.3. Debris falls.…………………………………………………………………..57
5.4.3.1. Occurrence of debris falls in the study area...……………………….57
5.4.4. Rock avalanches...…………………………………………………………....58
5.4.4.1. Occurrence of rock avalanche in the study area.....…………………58
5.4.5. Slump and rock fall.…………………………………………………………58
5.4.5.1. Occurrence of slump and rock fall in the study area..……………...59
5.5. Geographic distribution of mass movements in the study area.. ………………………59
5.5.1. Mass movements in Muzaffarabad and surrounding area. .…………………….60
5.5.2. Mass movements in Jhelum valley area...……………………………………….68
5.5.3. Mass movements in Neelum valley area...……………………………................71
5.6. Mass movement case studies..………………………………………………………….76
5.6.1. Previous studies..….……………………………………………………………..77
5.6.2. Hattian Bala rock avalanche..….………………………………………...............77
5.6.2.1. Introduction to the Hattian Bala rock avalanche case study..…..………77
5.6.2.2. Geological setting of the Hattian Bala rock avalanche..….. …………...78
5.6.2.3. Structural setting of the Hattian Bala rock avalanche........... .………….80
5.6.2.4. Description of the Hattian Bala rock avalanche……..………………….82
5.6.3. Langarpura rock fall..…………………………………………………………...88
5.6.3.1. Introduction to the Langarpura rock fall case study..….………………88
5.6.3.2. Geological setting of the Langarpura rock fall..…. …………………...89
5.6.3.3. Description of the Langarpura rock fall…...…………………………..90
5.6.4. Neelidandi rock fall.....…………………………………………………………93
5.6.4.1. Introduction to the Neelidandi rock fall case study.......………………93
5.6.4.2. Geological setting of the Neelidandi rock fall.…. .…………………...94
5.6.4.3. Description of the Neelidandi rock fall…..…………………………...95
5.6.5. Panjgran slump and rock fall...….……………………………………………101
5.6.5.1. Introduction to the Panjgran slump and rock fall case study.....…….101
5.6.5.2. Geological setting of the Panjgran slump and rock fall… .…………101
5.6.3.4. Description of the Panjgran slump and rock fall..…………………...102
List of Contents
xii
5.6.6. Conclusions based on the case histories..………..…………………………..106
6. Statistical analysis of the mass movement distribution triggered by the Kashmir
earthquake 2005…………………………………………………....................................108
6.1. Introduction…………………………………………………………………………..108
6.2. Methodology………………………………………………………………………….108
6.3. General mass movement distribution..…………………………………………………10
6.4. Mass movement concentration as function of distance to earthquake source.…….….112
6.4.1. Mass movement concentration in terms of distance from epicenter..……..…...114
6.4.2. Mass movement concentration in terms of distance from Muzaffarabad Fault..117
6.4.3. Mass movement concentration in terms of distance from hanging
wall and foot wall blocks of Muzaffarabad Fault ...…………………………...121
6.5. Mass movement concentration in terms of topographic parameters…………………..123
6.5.1. Mass movement concentration as function of slope steepness…………………124
6.5.2. Mass movement concentration as a function of slope aspect……………….….125
6.5.3. Mass movement concentration as function of elevation…………………….....127
6.6. Distribution of mass movements as function of geological units..…………………...129
6.7. Discussion and conclusions…..……………………………………………………….134
7. Empirical analysis of geometrical parameters of mass movements triggered by the
Kashmir earthquake 2005. ……………………………………………………………...137
7.1. Empirical models…..…………………………………………………………………137
7.2. Study background….….……………………………………………………………...137
7.3. Data source and methodology………….. ……………………………………………139
7.4. Types of considered mass movements .. ……………………………………………..144
7.5. Geometrical parameters considered for empirical analysis …………………………145
7.5.1. Fahrböschung angle…….………………………………………………………146
7.5.2. Shadow angle….………………………………………………………………..148
7.5.3. Talus slope angle…..……………………………………………………………148
7.6. Analysis of Kashmir earthquake 2005 mass movement data…………………………148
7.7. Results of analysis..……………………………………………………………………149
7.7.1. Relationship between the mass movement volume and Fahrböschung
angle for all types of mass movements. ..……....………………………………149
7.7.2. Relationship between the mass movement volume and Fahrböschung
angle for individual groups of mass movements. ..…………………………..151
List of Contents
xiii
7.7.3. Relationship between mass movement volume versus shadow
angle and talus slope angle……………………………………………………….154
7.7.4. Relationship between the Fahrböschung angle and the height of fall…..………157
7.7.5. Relationship between the height of fall (H) and the travel distance (L)..…… ...158
7.7.6. Relationship between surface area and volume of mass movements….… ..…..160
7.8. Comparison between international published data and own results… ………………..161
8. Conclusions and recommendations………………………………………………........166
8.1 Conclusions ………………………………………………………………………….166
8.2 Recommendations……………………………………………………………………167
References……………………………………………………………………………….169
Appendices……………………………………………………………………………….183
List of Figures
xiv
LIST OF FIGURES
Fig. 1.1 Map of the northern Pakistan and Kashmir region. The epicenter of Kashmir
earthquake 2005 was located in the Kashmir region (http: // www.drgeorgepc.com /
Earthquake2005Pakistan.html). The inset map shows the position of Pakistan within the
world map (http://www.worldatlas.com). …............................................................................ 2
Fig. 1.2 The investigated area is marked by the rectangular polygon, lies in Pakistani
Administrated Kashmir (PAK). The area was badly affected by the Kashmir earthquake
2005 in terms of life losses as well as earthquake induced mass movements (map of
Muzaffarabad district digitized and modified after the map from Planning and
Development department AJK, 2007). The inset map shows the location of Muzaffarabad
district in PAK. …….............................................................................................................. 5
Fig. 1.3 3D outlook view of the study area, affected by the Kashmir earthquake 2005. The
image is formed by SPOT-5 imagery over NASA SRTM (Shuttle Radar Topography
Mission) digital topography. …………………………………………….............................. 6
Fig. 1.4 Organization of PhD thesis structure………………………………………………. 8
Fig. 2.1 Map of Indian-Eurasian collision zone showing the Himalaya as a barrier between
the Tibetan Plateau and the plains of Indian plate (file:///E:
/Maps/JPEGMAP/File_Himalayas_ Map.htm). The inset map shows the location of Indian-
Eurasian collision zone in the world map. ……………....................................................... 10
Fig. 2.2 Tectonic map of the Himalayas showing the main tectonic zones (compiled after
Gansser, 1981; Windlay, 1983). …………………………………………………… 11
Fig. 2.3 The regional tectonic map shows the major tectonics features in northern Pakistan
(compiled after Wadia, 1931; Calkins et al., 1975; Baig and Lawrence, 1987; Greco, 1991;
Avouac et al., 2006 ; Baig, 2006 and Yeats et al., 2006). …………………………. 12
Fig. 2.4 Enlarged regional tectonic map of northern Pakistan. Sources used as in Fig. 2.3
MKT: Main Karakorum Thrust, MMT: Main Mantle Thrust, MCT: Main Central Thrust,
PT: Panjal Thrust, MBT: Main Boundary Thrust, HFT, Himalayan Frontal Thrust, JF:
Jhelum Fault. ……................................................................................................................ 16
Fig. 2.5 Structural map of the Hazara Kashmir Syntaxis (compiled and modified after
Wadia, 1931; Calkins et al., 1975; Baig and Lawrence, 1987; Greco, 1989). HKS: Hazara
Kashmir Syntaxis; MBT: Main Boundary Thrust; PT: Panjal Thrust; HFT: Himalayan
Frontal Thrust. ……………………………………………………………………………. 19
List of Figures
xv
Fig. 2.6 The Muzaffarabad Fault runs between the Muzaffarabad Formation and Murree
Formation in Makri area, Muzaffarabad city. ………………………………………………. 21
Fig. 3.1 Map showing the geological units and faults of Hazara Kashmir Syntaxis(compiled
after Wadia, 1931; Latif, 1970; Calkins et al., 1975; Baig and Lawrence, 1987; Greco,
1989; Hussain et al., 2004; Munir et al., 2006 and Kaneda et al., 2008). ……….................. 24
Fig. 3.2 Geological map of Jhelum valley, Neelum valley and Muzaffarabad city (compiled
and modified after Wadia, 1931; Latif, 1970; Calkins et al., 1975; Baig and Lawrence,
1987; Greco, 1989; Hussain et al., 2004, Munir et al., 2006 and Kaneda et al., 2008). … 25
Fig. 3.3 The slates of Hazara Formation are well exposed along Muzaffarabad–Mansehra
road in Muzaffarabad area. ……………………………………………………………… 27
Fig. 3.4 The highly crushed and sheared bedded Muzaffarabad Formation exposed at
Kamsar area, in the northeast of Muzaffarabad city. ……………………………………….. 29
Fig. 3.5 Margala Hill limestone exposed at Yadgar area, Muzaffarabad……..…………….. 32
Fig. 3.6 Fractured and jointed sandstone of the Miocene Murree Formation exposed in the
Jhelum valley area. ………………………………………………………………………….. 33
Fig. 3.7 The alluvial fan deposits near Chellah Bandi, Muzaffarabad………………………. 35
Fig. 3.8 Talus slope deposit at the base of cliff at Shahwi nala, Muzaffarabad…………….. 35
Fig. 4.1 Methodological steps used during the research work………………………………. 37
Fig. 4.2 SPOT-5 (A) and Quickbird (B) imagery show the mass movements triggered by
the Kashmir earthquake 2005 in Botha area Muzaffarbad. ………......................................... 39
Fig. 4.3 The mass movements interpreted from SPOT satellite imageries and field
investigation within the study area. …………………………………………………………. 40
Fig. 4.4 Examples of the mass movements triggered by the Kashmir earthquake 2005 in
NW Himalayan, Pakistan. A. Subri rock fall blocked the Jhelum valley road during the
earthquake, B. Rockslide of Saran area caused to damage the houses, C. Dehrian Saidan
rock fall on hills of Muzaffarabad city, D. Debris fall near the Pattika town, E. Author
measuring absolute horizontal distance of mass movement in field, and F. Fractured
sandstone in the Miocene Murree Formation at Makri, Muzaffarabad. …………………….. 42
Fig. 4.5 Mass movements triggered by the Kashmir earthquake 2005 data base interface
prepared in ArcGIS 9.3. ……................................................................................................... 44
Fig. 5.1 Map showing the distribution of mass movements triggered by the Kashmir
earthquake 2005, in Muzaffarabad and surrounding areas. The investigated area is marked
by a rectangular polygon (map of Muzaffarabad district digitized and modified after the
map from Planning and Development department AJK, 2007). ……………………………. 48
List of Figures
xvi
Fig. 5.2 Multiple shallow failures on very steep slopes in the Hanging Wall Block (HWB)
Muzaffarabad Fault (MF), northeast of Muzaffarabad city. The Muzaffarabad Formation
lies in the HWB and the Murree Formation in the Foot Wall Block (FWB). ………………. 51
Fig. 5.3 SPOT-5 (2.5 m) image showing the mass movements induced by the Kashmir
earthquake 2005 denudated the vegetation on the steep slopes in the north and northwestern
part of Muzaffarabad city, and around the Muzaffarabad Hills. …………............................. 52
Fig. 5.4 Map showing the distribution and the types of mass movements investigated
during field surveys in 2009 and 2010 for this study…........................................................... 54
Fig. 5.5 Transitional landslide in Botha area, in the northeast of Muzaffarbad city triggered
by the earthquake. ………………………………………………………………………...... 55
Fig. 5.6 The rock fall occurred in weathered shales and fractured sandstones of Miocene
Murree Formation near Batmang area along the main road of Neelum valley. This rock fall
blocked the Neelum valley road many days after the earthquake. ………………………….. 56
Fig. 5.7 Debris fall located in Dunga Kas Nala in the Neelum valley area triggered by the
Kashmir earthquake 2005. This debris fall was occurred at low altitude area along the
stream. ……………………………………………………………………………………….. 58
Fig. 5.8 Map showing the location of the 3 main study areas: 1. Muzaffarabad city 2.
Jhelum valley, 3. Neelum valley. …………......................................................................... 59
Fig. 5.9 Quickbird (0.6 m) image showing shallow and deep mass movements on steep
slopes of Muzaffarabad hills, around Muzaffarabad city. Out line shows the boundary of
the mass movements. ………................................................................................................... 61
Fig. 5.10 Mass movement distribution map of the Muzaffarabad city and surrounding
area…………………………………………………………………………………………… 62
Fig. 5.11 Types of mass movements in Muzaffarabad city and the surrounding area………. 63
Fig. 5.12 Mass movement failures on the steeper slopes of carbonate rocks in the northern
and northwestern part of Muzaffarabad city ………………………………………………… 66
Fig. 5.13 Shallow slope failures on the steep slope of Muzaffarabad Hills in dolomitic
limestone. Most of the activated material was the former talus cover. …………………....... 67
Fig. 5.14 Mass movements in the southwest of Muzaffarabad city reactivated by the
Kashmir earthquake 2005. ………………………………………………………………....... 68
Fig. 5.15 Mass movement distribution map of the Jhelum valley area……………………… 69
Fig. 5.16 Debris fall occurred in the red clay of the Miocene Murree Formation in Jhelum
valley area triggered by the Kashmir earthquake 2005, additionally caused by undercutting
of the slope for road construction. ………………………………………………………....... 70
List of Figures
xvii
Fig. 5.17 Mass movement distribution map of the Neelum valley area……………………... 73
Fig. 5.18 A view of the Nauseri rock fall close to the MBT in the Neelum valley area…….. 74
Fig. 5.19 Map showing the locations of the case studies described in text. (map of
Muzaffarabad district digitized and modified after the map from Planning and
Development department AJK, 2007). ……………................................................................ 76
Fig. 5.20 SPOT-5 image of the Hattian Bala rock avalanche. Outline shows the boundary
of the mass movement and blockage of the Karli and Tung tributaries of the Jhelum
river………………………………………………………………………………………... 78
Fig. 5.21 Geological map of the Hattian Bala and surrounding area (compiled and modified
after GSP, 2008). ……………………………………………………………………………. 79
Fig. 5.22 Structural map showing the southeast plunging synclinal structural failure of
Hattian Bala rock avalanche. ………………………………………………………………... 81
Fig. 5.23 Geotechnical cross profiles (5.23a, 5.23b) showing the pre-earthquake and post-
earthquake situation. Note: the rock avalanche perfectly follows the pre-existing structure
of the Danna and Dandbeh synclines. The third profile (5.23c) is showing the maximum
deposit thickness of the Hattian Bala rock avalanche. ………………………........................ 82
Fig. 5.24 View of the Hattian Bala rock avalanche structurally controlled by southeast
plunging Danna syncline. Photo facing northwest…………………………………………... 83.
Fig. 5.25 Multiple extensional ground cracks oriented northeast-southwest on the crown of
the Hattian Bala rock avalanche. Photo facing northwest. ………………………………….. 84
Fig. 5.26 Map of the Hattian Bala rock avalanche from 2005. Note: situation of old rock
slide and it position is derived from Schneider (2008). …….................................................. 85
Fig. 5.27 Geotechnical map of the Hattian Bala rock avalanche and the locations of
geological longitudinal and geotechnical cross profiles. Frequent GPS measurements were
performed during field work to mapped the geotechnical details. ………………………….. 86
Fig. 5.28 Longitudinal geotechnical NW-SE profile showing pre-earthquake landscape and
the geotechnical situation after the rock avalanche. Note: the mass movement is parallel to
the southeast orientated synclinal plunge direction and the slip surface follows in many
parts the dip direction of the bedding. The mass movement abuts and accumulates against
the right steep wall of the former Karli valley. …………………………………................. 87
Fig. 5.29 The deposit of the Hattian Bala rock avalanche is mainly composed of angular
rock fragments. ……………………………………………………………………………… 88
List of Figures
xviii
Fig. 5.30 SPOT-5 image of the Langarpura rock fall in the Jhelum valley area. The red line
shows the location of Muzaffarabad Fault passing through this area. Outline shows the
boundary of the mass movement. …………………………………………………………… 89
Fig. 5.31 The escarpment failure of the Langarpura rock fall on the hanging wall block
(HWB) of the Muzaffarabad Fault. Note: The significant topographic front formed by the
hanging wall block of the Muzaffarabad Fault. The foot wall block (FWB) has been eroded
by river undercutting before the reactivation of Langarpura rock fall. Photo looking
northeast. …………………………………………………………………………………….. 90
Fig. 5.32 The geotechnical map of the Langarpura rock fall and the location of the
geological longitudinal profile shown in Fig. 5.33. ……………............................................. 92
Fig. 5.33 Geological longitudinal profile of the Langarpura rock fall. Location of the
profile is shown in Fig. 5.32. ………………………………………………………………... 93
Fig. 5.34 SPOT-5 image of the Neelidandi rock fall in north of Muzaffarabad city. The
rock fall blocked the Neelum river for 5 hours immediately after the earthquake. Outline
shows the boundary of the mass movement. ……………………………………………....... 94
Fig. 5.35 The very steep scarp of Neelidandi rock fall in the hanging wall block of
Muzaffarabad Fault (MF). The scarp is formed in the highly sheared cherty dolomite of
Muzaffarabad Formation. The Muzaffarabad Formation lies in the hanging wall block and
Murree Formation in the foot wall block. Photo looking southeast. …………....................... 96
Fig. 5.36 Northwest dipping and oblique tension cracks on the crown of the Neelidandi
rock fall.The cracks are formed due to earthquake ground shaking and stress release behind
the new steep scarp. These cracks can cause a collapse of material behind the scarp during
monsoon rains and earthquake shocks. Photo looking northeast………………………….... 97
Fig. 5.37 Geotechnical map of the Neelidandi rock fall and the location of the geological
longitudinal profile shown in Fig. 5.39. …………………………………………………….. 98
Fig. 5.38 The hanging wall block of Muzaffarabad Fault is deformed into active hanging
wall anticline (Baig et al. 2008). Note: total destruction of houses (Earthquake intensity XI)
is due to strong earthquake ground shaking along Muzaffarabad Fault. Photo looking
northeast…………………………………………………………………………................... 99
Fig. 5.39 Geological longitudinal profile of the Neelidandi rock fall. Location of the profile
is shown in Fig. 5.37. ………………………………………………………………………... 100
Fig. 5.40 SPOT-5 image showing the location and boundary of the Panjgran slump and
rock fall occurred in the northeast of Muzaffarabad, in the Neelum valley area. Outline
shows the boundary of the mass movement. ………………………………………………... 102
List of Figures
xix
Fig. 5.41 An overview of the Panjgran slump and rock fall which occurred close to the
epicentral region of the Kashmir earthquake 2005. Note: the position on an undercut slope
of Neelum river. ……............................................................................................................... 103
Fig. 5.42 Geotechnical map of the Panjgran slump and rock fall and the location of the
geological longitudinal profile shown in Fig. 5.43. …………………………………………. 104
Fig. 5.43 Geological longitudinal profile of the Panjgran slump and rock fall. Location of
the profile is shown in Fig. 5.42. ……………………………………………………………. 105
Fig. 6.1 Mass movement distribution map of the Kashmir earthquake 2005 in the vicinity
of Muzaffarabad, Jhelum and Neelum valleys. The investigated area is marked by a
rectangular polygon. ………………………………………………………………………… 109
Fig. 6.2 Map showing the geological setting of the study area affected by the mass
movements (compiled after Wadia, 1931; Calkins et al., 1975; Baig and Lawrence, 1987;
Greco, 1989; Hussain et al., 2004, Munir et al., 2006 and Kaneda et al., 2008). …………... 112
Figs. 6.3 (a) Mass movement distribution around the epicenter within the whole study area,
(b) Mass movement distribution around the Muzaffarabad Fault within the whole study
area (c) Mass movement distribution involving an area of 10 km around the epicenter, (d)
Mass movement distribution involving an area of 10 km around the rupture of the
Muzaffarabad Fault. …………………………………………………………………………. 113
Fig. 6.4 Number, area of mass movement and mass movement concentration defined in
terms of distance from the epicenter. ………………………………………………………... 115
Figs. 6.5 (a) Relationship between the mass movement concentration and the distance from
the epicenter for over all data, (b) Relationship between the mass movement concentration
and the distance 10 km away from the epicenter. ………........................................................ 116
Fig. 6.6 Number, area of mass movement and mass movement concentration defined in
terms of distance from the Muzaffarabad Fault. …………………………………………….. 118
Figs. 6.7 (a) Relationship between the mass movement concentration and the distance from
the Muzaffarabad Fault for over all data, (b) Relationship between the mass movement
concentration and the distance 10 km away from the Muzaffarabad Fault. ………………… 120
Fig. 6.8 Mass movement distribution along hanging and foot wall blocks of the reactivated
Muzaffarabad Fault around 7 km distance away from the Muzaffarabad fault rupture. 121
Fig. 6.9 Number, area of mass movement and concentration of mass movements on the
hanging wall block (left) and foot wall block (right) of the Muzaffarabad Fault. …………... 122
Fig. 6.10 (a) Slope angle distribution of the study area (b) slope aspect distribution of
thestudy area (c) DEM of the study area. …...................................................................... 123
List of Figures
xx
Fig. 6.11 Number, area of mass movement and the relationship between the mass
movement concentration and slope gradient categories. ……………………………………. 125
Fig. 6.12 Number, area of mass movement and the relationship of mass movement
concentration and slope aspect. …………......................................................................... 127
Fig. 6.13 Number, area of mass movement and the relationship between mass movement
concentration and elevation of the study area………………………………………………... 128
Fig. 6.14 The percentage of mass movement occurrence in terms of geological units……… 131
Fig. 6.15 The percentage of mass movement area in terms of geological units……………... 132
Fig. 6.16 Number, area of mass movement and the slope failure in different rock types
defined in terms of mass movement concentration. ……........................................................ 133
Fig. 7.1 Location of the mass movement events (represented by triangles), identified for
empirical analysis of geometrical parameters of rock falls, triggered by the Kashmir
earthquake 2005, in the vicinity of Muzaffarabad city, Jhelum valley and Neelum valley.
The investigated area is marked by the rectangular polygon. …………………………......... 138
Fig. 7.2 Examples of rock falls triggered by the Kashmir earthquake 2005, considered for
empirical analysis. a) Battalian rock fall in the Jhelum valley, b) Makri rock fall in the
Muzaffarabad city, c) Nauseri rock fall in the Neelum valley, d) Devlian rock fall in
Neelum valley. ………............................................................................................................. 140
Fig. 7.3 Sketch of mass movement source point, falling mass and deposit. Definitions of
parameters used in the present analysis are explained in text. Sketch is modified from
Evans and Hungr (1993) and Copons et al., (2009). ………………………………………... 146
Fig. 7.4 Relationship between log tangent of the Fahrböschung and log mass movement
volume for all investigated rock and debris falls triggered by the Kashmir earthquake
2005……………………………………………………………………………………… 150
Fig. 7.5 Fahrböschung angles and travel distances of individual groups of rock and debris
falls triggered by the Kashmir earthquake 2005. ……………………………………………. 152
Fig. 7.6 Relationship between tangent of the Fahrböschung and volume for very large rock
falls, large rock falls, medium debris falls, small debris falls, and very small debris
falls………………………………………………………………………………………. 153
Fig. 7.7 Relationship between log tangent of the shadow angle and volume of 55 selected
events for all types of rock and debris falls triggered by the Kashmir earthquake 2005. . 156
Fig. 7.8 Relationship between log of the tangent of the talus slope angle and volume of 41
selected events for all types of rock and debris falls triggered by the Kashmir earthquake
2005. ……………………………………………………………………………….. ……….. 156
List of Figures
xxi
Fig. 7.9 Relationship between tangent of the Fahrböschung and the height of fall for all
types of mass movements triggered by the Kashmir earthquake 2005. ……................... 157
Fig. 7.10 Relationship between the log height of the fall (H) and the log travel distance (L)
for 20 all types of rock falls triggered by the Kashmir earthquake 2005. ……………… 158
Fig. 7.11 Relationship between the log height of fall (H) and the log travel distance (L) for
83 all types of debris falls triggered by the Kashmir earthquake 2005. …….................... 159
Fig. 7.12 Relationship between tangent of the Fahrböschung and the travel distance (L) for
all types of rock and debris falls triggered by the Kashmir earthquake
2005……………………………………………………………………………………... 160.
Fig. 7.13 Relationship between log surface area and log mass movement volume for all
types of mass movements triggered by the Kashmir earthquake 2005. …………………….. 161
Fig. 7.14 Relationship between log tangent of the Fahrböschung and log mass movement
volume. Comparison between the data of Scheidegger (1973), Erismann & Abele (2001),
Copons et al., (2009) and the dataset of Kashmir earthquake 2005. ………………………... 163
Fig. 7.15 Relationship between log tangent of the Fahrböschung and log mass movement
volume. Comparison between international data and own data set (Kashmir earthquake
2005 data)……………………………………………………………………………………. 165
List of Tables
xxii
LIST OF TABLES
Table 3.1 Geology of the earthquake affected area……………………………………… 26
Table 5.1 Mass movement classification (Varnes, 1978)………………………………… 49
Table 5.2 Mass movement classification (Cruden and Varnes, 1996)…………………… 50
Table 5.3 Types of mass movements examined in the field (Classification after Varnes
1978). …………………………………………………………………………………… 53
Table 5.4 Type, length, width, area and volume of mass movements in Muzaffarabad city
and the surrounding area. ……............................................................................................. 64
Table 5.5 Type, length, width, area and volume of mass movement distribution in Jhelum
valley area. ………………………………………………………………………………… 71
Table 5.6 Type, length, width, area and volume of mass movement distribution in Neelum
valley area. ………………………………………………………………………………… 74
Table 5.7 Structural data of Danna syncline, Hattian Bala rock avalanche………………… 80
Table 5.8 Geometric characteristics of the Hattian Bala rock avalanche triggered by the
Kashmir earthquake 2005, in northern Pakistan. ………………………………………….. 86
Table 5.9 Geometric characteristics of the Langarpura rock fall triggered by the
Kashmir earthquake 2005. ………………........................................................................... 91
Table 5.10 Geometric characteristics of the Neelidandi rock falls triggered
by the Kashmir earthquake 2005. ………………………………………………………… 99
Table 5.11 Geometric characteristics of the Panjgran slump and rock fall triggered by the
Kashmir earthquake 2005. ……………………………………………………………….. 104
Table 6.1 The relationship between mass movement concentration and slope steepness
within the study area of Kashmir earthquake 2005. …………………………………….... 124
Table 6.2 The relationship between mass movement concentration and slope aspect
within the study area of Kashmir earthquake 2005. …………............................................ 126
Table 6.3 The relationship between mass movement concentration and elevation
within the study area of the Kashmir earthquake 2005. ………………………………….. 128
Table 6.4 Geological formations, lithological description, age, percentage of mass
movement, percentage of surface area for geological units and mass movement
concentration. …………………………………………………………………………….. 130
Table 7.1 Geometrical data of 103 mass movement events triggered by the Kashmir
earthquake 2005, in the vicinity of Muzaffarabad city, the Jhelum valley and the Neelum
valley, in northern Pakistan. ………………………………………………………………. 141
List of Tables
xxiii
Table 7.2 Rock fall classification based on a volumetric nomenclature
(after Varnes, 1978 and Fell, 1994). ………………………………………………………. 144
Table 7.3 Classification of fall-types based on volume and number of mass movements… 145
Table 7.4 Results of the linear regression analysis of tangent of the Fahrböschung versus
the mass movement volume of individual rock fall groups. ……………………………… 154
xxiv
ABBREVIATIONS
AJK Azad Jammu and Kashmir
ASTER Advanced Spaceborne Thermal Emission and Reflection Radiometer
AKMIDC Azad Kashmir Mineral and Industrial Development Corporation
ASL Above Sea Level
BGR Bundesanstalt für Geowissenschaften und Rohstoffe
DEM Digital Elevation Model
ERRA Earthquake Reconstruction and Rehabilitation Authority
GSP Geological Survey of Pakistan
GPS Global Positioning System
HKS Hazara Kashmir Syntaxis
HFT Himalayan Frontal Thrust
IAK Indian Administrated Kashmir
ISZ Indus Tsangpo Suture Zone
IKSZ Indus Kohistan Seismic Zone
JICA Japan International Corporation Agency
JSCE Japan Society of Civil Engineers
JF Jhelum Fault
LoC Line of Control
NWFP North West Frontier Province
NESPAK National Engineering Services Pakistan
MBT Main Boundary Thrust
MKT Main Karakorum Thrust
MCT Main Central Thrust
MFT Main Frontal Thrust
MF Muzaffarabad Fault
PAK Pakistani Administrated Kashmir
PT Panjal Thrust
SRT Salt Range Thrust
SRTM Shuttle Radar Topography Mission
SPOT Satellite Pour l’Observation de la Terre
UTC Universal Coordinated Time
USGS United States Geological Survey
xxv
Peer reviewed publication
1. Basharat, M., Rohn, J., Baig, M. S., Ehret, D. (2012): The lithological and structural
control of Hattian Bala rock avalanche triggered by the Kashmir earthquake 2005, Sub-
Himalayas, Northern Pakistan. Journal of Earth Science. Vol. 23, No. 2, p. 213-224,
DOI: 10.1007/s12583-012-0248-3
Abstracts
1. Basharat, M., Rohn, J., Baig, M. S., Ehret, D. (2010): The Pattern, Geological
Parameters and Distribution of Mass Movements triggered by Kashmir Earthquake 2005 in
Northern Pakistan Geophysical Research Abstracts, Vol. 12, EGU 2010-7756, 2010 EGU
General Assembly, 2010.
2. Basharat, M., Rohn, J., Moser, M. (2010): Statistical Analysis Of Mass Movements
Triggered By Kashmir Earthquake 2005 And Their Run-Out Distance. American
Geophysical Union Fall Meeting 2010, abstract # NH31A-1331
Chapter 1: Introduction
1
Introduction
1.1. Backdrop
A big number of disastrous earthquakes occurred during the last ten years in mountain regions like
in El Salvador (2001), Pakistan (2005), China (2008) and Haiti (2010) triggering thousands of
mass movements throughout the areas. The area affected by the mass movements is related to the
earthquake magnitude, epicenter, focal depth, geological conditions and ground motion
characteristics (Keefer, 1984, 2002). Moreover, mass movements occur in a variety of geological
environments ranging from steep rock slopes to gentle slopes (Keefer, 1984). These mass
movements caused horrendous loss of life, great human sufferings and miseries, and wide spread
loss of property and infrastructure. Sometimes the damage caused by the mass movements
exceeded the damage directly related to the earthquakes (Schuster and Fleming, 1986). The most
disastrous mass movements during earthquakes have claimed as many as 100,000 lives in 1920
Haiyuan earthquake China (Li and Wang, 1992).
The Himalayan mountain chain is one of the most seismically active regions in the world and
causes frequent earthquakes and coseismic mass movements. The disastrous Kashmir earthquake
(magnitude Mw. 7.6) occurred on 8th October 2005 (3:50:40 Universal Coordinated Time (UTC)),
in the northwestern part of the Himalayan range in northern Pakistan (Fig. 1.1). Its epicenter was
located (34º 29′ 35″ N and 73º 37′ 44″ E) ~18 km northeast of the city of Muzaffarabad at a focal
depth of 26 km in Kashmir region (USGS, 2006). The devastating earthquake was the worst
natural disaster in the last 100 years in Kashmir (Bendick et al., 2007). The earthquake killed more
than 73,000 people, 69,000 people were injured and 2.8 million people were left homeless (official
sources). This earthquake triggered a number of mass movements that caused about 30 % of all
fatalities directly or indirectly and huge economic losses (Petley et al., 2006). The economic losses
by this earthquake including reconstruction and rehabilitation were estimated to be 5.2 billion US
$ (Asian Development Bank and World Bank, 2005).
Due to a lack of awareness in Pakistan, the impact of mass movements was not ascribed a big
importance to the population as well as to the government. The knowledge regarding mass
movement events occurring in past was limited or even unknown in the area. Before the
earthquake of 8th October 2005, any kind of systematic scientific investigation on landslides was
not carried out in this region. As a result, the people and government paid high price in terms of
human life, loss of land and in the relocation and reconstruction of entire villages. The impact of
mass movements on human society is increasing. The reason for this increase was primarily
human related and includes encroachment of human population at steep slopes and deforestation.
Chapter 1: Introduction
2
Fig. 1.1 Map of the Pakistan and Kashmir region. The epicenter of Kashmir earthquake 2005 was
located in the Kashmir region, northern part of Pakistan (http: // www.drgeorgepc.com /
Earthquake 2005 Pakistan.html). The inset map shows the position of Pakistan within the world
map (http: // www.worldatlas.com).
After the devastating effect of mass movements triggered by the Kashmir earthquake 2005,
significant efforts has been made by the scientific community, local and international
organizations as well as by the institutes and the universities to identify and characterize the mass
movements and to improve the knowledge and understanding of the process. As a part of this
commitment, it is mandatory to understand the mechanism, distribution, characteristics and
behaviour of mass movements in order to assess and mitigate the geological risk pose to the
potential affected area.
Chapter 1: Introduction
3
1.2. Literature review
After the Kashmir earthquake 2005, the distribution and characteristics of mass movements have
been investigated by several researchers throughout the affected area (Fujiwara et al., 2006; Harp
and Crone, 2006; Petley et al., 2006; Dunning et al., 2007; Sato et al., 2007; Kamp et al., 2008;
Owen et al., 2008; Schneider, 2008; Champati Ray et al., 2009 Saba et al., 2010 and Peduzzi,
2010). Sato et al. (2007) initially studied the distribution of mass movements triggered by the
Kashmir earthquake 2005. Using remote sensing data, they identified 2,424 mass movements in an
affected area of more than 7,500 km2. They described that the mass movements mostly occurred
close to the active fault along the hanging wall block. However, they did not explain the
relationship between mass movements and geological parameters. Petley et al. (2006) reported the
number of fatalities associated with mass movements in this earthquake. They estimated that more
than 30% of the people died directly or indirectly as a result of mass movements. They also
examined the wide spread occurrence of cracked slopes which threaten the local inhabitants
around Muzaffarabad. Owen et al. (2008) studied the mass movements in an area of 750 km2 near
Muzaffarabad and Balakot affected by the Kashmir earthquake 2005. They identified 1,293 mass
movements at 174 locations and developed a first mass movement inventory in the field
identifying the types of slope failure. However, they did not study the detailed geological and
structural characterization. Kamp et al. (2008) produced a susceptibility map for the region and
pointed out that the mass movements are concentrated in specific zones associated with event
controlling parameters like deforestation and road construction. These factors contributed
significantly to the frequency of mass movements during and after the earthquake. Saba et al.
(2010) carried out a preliminary study of the mass movements using pre- and post-earthquake
Quickbird, IKONOS, Satellite Pour l’Observation de la Terre (SPOT-5), and WorldView01
imageries. They identified 158 mass movements along Balakot-Bagh fault line within an area of 36
km2. They reported that the mass movement activity decreased within two years after the
earthquake.
However, limited work has been done so far, where mass movement distribution was statistically
analyzed. Moreover, physical characteristics of the mass movements triggered by the Kashmir
earthquake 2005 were not discussed earlier. Keeping in view the significance of the mass
movement problems in the area, the present study was undertaken regarding the distribution,
characteristics and behaviour of the mass movements triggered by the Kashmir earthquake 2005.
Chapter 1: Introduction
4
1.3. Research area
The inherently unstable nature of mountain areas in the northern part of Pakistan is well
recognized. Seismically active zones, steep slopes, disadvantageous geological conditions and
intense monsoon rains make this part of the Himalayas to one of the most hazard-prone areas in
the world.
The investigated area lies in Pakistani Administrated Kashmir (PAK) in the northwestern part of
the Himalayas in Kashmir region. The study was conducted within an area of approximately 1,299
km2, principally located in the vicinity of Muzaffarabad city, Jhelum valley, and Neelum valley
(Fig. 1.2). Politically, the area is disputed between India and Pakistan since their independence in
1947. The eastern boundary of PAK is still demarcated by the Line of Control (LoC), as
established by the two states (Pakistan and India). Muzaffarabad is the capital of PAK also named
as Azad Jammu and Kashmir (AJK) and lies only 135 km away from the capital of Pakistan,
Islamabad. It is located on the bank of the Jhelum and Neelum rivers. Muzaffarbad city is the main
gateway to enter the Jhelum and Neelum valleys up to a distance of only some few km from the
LoC.
Topography of the area is mainly mountainous with terraces and deep incised valleys (Fig. 1.3).
The ridges and valleys generally trend southeast-northwest parallel to the regional geological
structures. The Neelum river flows across this trend from north to south. However, the Jhelum
river flows parallel to southeast northwest regional strike of the Jhelum valley. In addition, the
Jhelum river joins the Neelum river near Muzaffarabad city and turns south. Generally, the relief
of the area is high in the north-northeastern and southeastern parts. To the north the topography is
characterized by steep mountains with elevations up to about 4,500 m asl. Whereas, to the south
the area is bounded by low hills of Murree and Abbotabad ranges at elevations of about 550 m asl
at the river bed.
The study area lies in the subtropical highland climate region within the reaching of strong
monsoon. In Muzaffarabad (about 700 m asl), the minimum and maximum temperature ranges
from -2.6 to 45.2 Cº (Planning and Development department AJK, 2010). However, the variation
in temperature is high towards the north of Muzaffarabad city in the Neelum valley area. During
the summer local rain showers are common and in winter the precipitation is mostly in the form of
snow above about 1,500 m asl. The average annual precipitation during the whole year is 1,300
mm (Planning and Development department AJK, 2010). As a result, severe flooding along the
rivers and streams are very common during the monsoon season.
Chapter 1: Introduction
5
Fig. 1.2 The investigated area is marked by the rectangular polygon, lies in Pakistani
Administrated Kashmir (PAK). The area was badly affected by the Kashmir earthquake 2005 in
terms of life losses as well as earthquake induced mass movements (map of Muzaffarabad district
digitized and modified after the map from Planning and Development department AJK, 2007). The
inset map shows the location of Muzaffarabad district in PAK.
The choice of the study area was determined by various considerations:
1. The disastrous Kashmir earthquake 2005 caused severe damage in PAK and North West
Frontier Province (NWFP) in Pakistan. In PAK, three main districts namely Muzaffarabad,
Bagh and Poonch were badly affected during the earthquake. Muzaffarabad was heavily
Chapter 1: Introduction
6
struck in terms of casualities, property losses, and destruction of the infrastructure. More
than 70 % of the total mass movements in PAK triggered by the earthquake occurred in the
Muzaffarabad district.
2. The study area lies along the rupture zone of the reactivated Muzaffarabad Fault, Main
Boundary Thrust (MBT) and close to the epicenter region. It was the area most affected by
mass movements during the Kashmir earthquake 2005.
3. The study area was chosen where mass movements are a significant problem and represent
a threat to the inhabitants of the area in terms of life losses and property.
4. In many parts of the study area, the mass movement problems have been increased by
human activities such as urban development, construction of roads on steep slopes and
deforestation.
Fig. 1.3 3D outlook view of the study area, affected by the Kashmir earthquake 2005. The image is
formed by SPOT-5 imagery over NASA SRTM (Shuttle Radar Topography Mission) digital
topography.
1.4. Aim of the study
The principal aim of this research was to improve the knowledge and understanding of the
distribution, characteristics and behaviour of the mass movements triggered by the Kashmir
earthquake 2005. The secondary aim was to develop a data base inventory and methodology that
Chapter 1: Introduction
7
can be used generally in developing countries. The basic technical information provided by this
study can be used by the national scientific community to produce hazard and risk assessment
maps for future planning.
The more specifically the research objectives were:
1. Identification and classification of mass movements triggered by the Kashmir earthquake
2005.
2. To produce mass movement inventory maps of Muzaffarabad city and the areas around
Jhelum valley and Neelum valley after the Kashmir earthquake 2005.
3. Investigation of large scale mass movements as case studies.
4. To understand distribution pattern of mass movement triggered by the Kashmir earthquake
2005.
5. To understand relationship between various geometrical parameters of mass movements
triggered by the Kashmir earthquake 2005.
1.5. Thesis structure
The thesis structure is organized into three parts as presented in Fig. 1.4. The first part of this
thesis contains the chapters of regional tectonics, geological setting, and methodology. In this part,
an overview of tectonics and geological setting of the study area are described with focus on active
tectonic features (Chapter 2), and lithological characteristics of different rock units (Chapter 3)
which affect the triggering of mass movements during the earthquake. In addition, the general
methodology (Chapter 4) used for this study are discussed.
The second part of this thesis is focusing on the mass movements triggered by the Kashmir
earthquake 2005 (Chapter 5). In this chapter, an attempt has been made to describe the types of
mass movements induced by the earthquake. In addition, the terminologies adopted for mass
movement classification and used for this work are explained. Moreover, mass movement
inventory maps were prepared using the information obtained during field investigations and
SPOT images. Furthermore, the investigations of four case studies (large scale mass movements)
have been discussed in terms of geological, structural, geotechnical and geometrical
characteristics.
The third part of the thesis focuses on the statistical analysis of the mass movement distribution
(Chapter 6) and the empirical analysis of geometrical parameters of the mass movements
(Chapter 7). The distribution of mass movements was analyzed statistically using regression
analysis, which allowed us to understand how the occurrence of mass movements correlates with
the distance from the earthquake source, geological conditions, and topographical parameters.
Chapter 1: Introduction
8
Likewise, empirical approaches were used to analyze the geometrical parameters of the mass
movements in order to determine the empirical relationship among various parameters on mass
movements travel distance.
Finally, the results are synthesized and interpreted in conclusions. The recommendations for future
work are made.
Fig. 1.4 Organization of PhD thesis structure.
Introduction
Regional Tectonics Geological Setting
Methodology
Mass movements triggered by the
Kashmir earthquake 2005
Statistical analysis
Empirical analysis
Conclusion
Part 1
Part 2 Part 3
Chapter 2: Regional tectonics
9
Regional tectonics
2.1. Tectonics of Himalayas
The world highest mountain chain, about 2,500 km long, 160-400 km wide, the Himalayas is
sandwiched between Eurasian plate and Indian plate. The Himalayan rugged mountain ranges,
extending from Pakistan in the northwest to India, Nepal and Bhutan in the northeast. It forms a
barrier between the Tibetan Plateau to the north and the plains of the Indian plate to the south (Fig.
2.1).
The Himalayan mountain belt was the result of continent-continent collision between Indian and
Eurasian plates. The Indian plate was separated from the Gondwana about 130 million years (Ma)
ago. In middle Cretaceous (80 Ma), Indian plate moved towards northwest, away from
Madagascar (Powell, 1979). As a result, the Newtethys between the Indian plate in the south and
Asian plate in the north started shrinking. This shrinking and continental drift was facilitated by
the consumption of Newtethys. During the closure of Newtethys, the Kohistan Island Arc
developed during late Jurassic to Cretaceous (Treloar and Izatt, 1993). The collisional boundary
between Eurasian plate and Kohistan Island Arc is referred to as the Main Karakorum Thrust
(MKT) (Tahirkheli, 1979). The collisional event began at 50-55 Ma (Powell, 1979; Patriate and
Achache, 1984) which is also supported by the fact that the Indian plate was rapidly drifting at a
rate of 130-150 millimeter per year northwards and collided with the Eurasian plate (Powell,
1979). The Indian plate from 50 Ma to present seemed to have moved northwards at much slower
rates of 40-60 millimeter per year (Powell, 1979). The abrupt slow moving rate is a result of the
Indian and Eurasian collision during the Early Tertiary (LeFort, 1975; Molnar and Tapponier,
1975).
The Kohistan Island Arc was docked onto the Eurasian plate in the north. The collision between
the Indian plate and the Kohistan Island Arc occurred during Eocene and is responsible for the
uplift of Himalayas (Molnar and Tapponier, 1975). The boundary between Indian plate and
Kohistan Island Arc is marked by the Main Mantle Thrust (MMT) (Tahirkheli, 1979). The south
migration of Himalayan deformation is represented by the MBT. The southern boundary of outer
Himalayas is believed to be tectonic boundary at certain places and called the Salt Range Thrust
(SRT) in Pakistan.
In the Indian Himalayas, northern collision zone has been identified as Indus Tsangpo Suture Zone
(ISZ), Main Central Thrust (MCT), MBT and Himalayan Frontal Thrust or Main Frontal Thrust
(HFT or MFT; Fig. 2.2). The ISZ extends west and bifurcates in Pakistan into two major
structures, the MMT and the MKT (Tahirkheli, 1979).
Chapter 2: Regional tectonics
10
Fig. 2.1 Map of Indian-Eurasian collision zone showing the Himalaya as a barrier between the
Tibetan Plateau and the plains of Indian plate (file:///E:/Maps/JPEGMAP/File_Himalayas _ Map.
htm). The inset map shows the location of Indian-Eurasian collision zone in the world map.
The northern suture MKT separates the intrusive and high grade metamorphic rocks of the
Eurasian plate from the Kohistan Island Arc. Whereas, the southern suture MMT separate the
Kohistan Island Arc from the hinterland of the Indian plate. The tectonic collision zone in northern
Pakistan has been subdivided by the MKT, MMT, MBT, SRT and HFT or Muzaffarabad Fault
(Tahirkhali, 1982; Farah et al., 1984; Yeats and Lawrence, 1984; Baig and Lawrence, 1987; Fig.
2.3).
Chapter 2: Regional tectonics
11
Fig. 2.2 Tectonic map of the Himalayas showing the main tectonic zones (compiled after Gansser,
1981; Windlay, 1983).
2.2. Tectonic setting of the NW Himalayas of Pakistan
In present tectonic setting, Pakistan lies on the northwestern corner of the Indian plate (Fig. 2.1). It
represents the part of the Tertiary convergence between Indian and Eurasian plates. In northern
Pakistan, Himalayas have four major subdivisions (Farah et al., 1984; Yeats and Lawrence, 1984).
These are from north to south; 1. Karakoram Ranges and Hindukush Ranges, 2. The Kohistan
Island Arc lies between the MKT and MMT (Tahirkheli, 1982; Farah et al., 1984), 3. The low
ranges of Swat, Hazara and Kashmir, located between the MMT and MBT, 4. Salt Range and
Potwar Plateau to the south of MBT represented the foreland fold and thrust belt in Sub Himalaya
(Figs. 2.2 and 2.3).
The Himalayan collision zone in east of Kashmir is tectonically subdivided by Gansser (1964 and
1981) from north to south into the Higher Himalayas, Lesser Himalayas and Sub Himalayas based
on structural, stratigraphic and morphological criteria.
Chapter 2: Regional tectonics
12
Fig. 2.3 The regional tectonic map shows the major tectonics features in northern Pakistan
(compiled after Wadia, 1931; Calkins et al., 1975; Baig and Lawrence, 1987; Greco, 1991; Baig,
2006 and Yeats et al., 2006).
Chapter 2: Regional tectonics
13
The MCT demarcates the Higher Himalayas from the Lesser Himalayas (Fig. 2.3). The MCT is
well developed in eastern and central Himalayas but its extension beyond Kaghan is still
controversial in northern Pakistan due to lack of evidence. The rock units of Higher Himalayas
represent the part of Indian basement and cover rocks (Greco, 1989). The Lesser Himalayan
sequence, locally poorly exposed in PAK, and Indian Administrated Kashmir (IAK) (DiPietro and
Pogue, 2004). The Lesser Himalayas are demarcated from the Sub Himalayas by the MBT (Fig.
2.3). The Lesser Himalayas comprises largely metasedimentary and sedimentary rock of
Precambrian to Tertiary age (Greco, 1989). The Sub Himalayas consists of the northern part of the
Indian shield, which has been covered by Tertiary molasse dominated by the Murree Formation
(Greco, 1989). The Sub Himalayas of northern Pakistan are defined by the Potwar-Kohat Plateau
comprising thick succession of Miocene-recent molasse sediments. South of the MBT in Pakistan,
the Sub Himalayas reaches its greatest breadth in the entire Himalayas. Clastic deposit of the
Miocene Murree Formation and overlying Siwalik Group of Miocene-Pleistocene age (Burbank
and Raynolds, 1984) lies in the Potwar Plateau. At the southern edge rises the Salt Range, in which
Eocambrian evaporates of the Salt Range Formation are overlain by Cambrian, Permain and
Mesozoic strata with relation to Gondwana (Gee, 1989). Most of the strata within Hazara Kashmir
Syntaxis (HKS) are Murree Formation, which contains red bed of sandstone, mudstone, shale and
claystone of early Miocene age that have developed slaty cleavage (Bossart et al., 1988).
Another classification by Coward et al., (1988) divided the Himalayas into internal (hinterland)
and external (foreland) zones on regional basis (Fig. 2.3). The internal zone to the south is
bounded by MMT, and consists of crystalline rocks of Naran, Upper Kashmir, Hazara, Besham
and Swat areas, whereas, the external zone, which in essence is a foreland thrust-fold belt,
comprises sedimentary rocks of low hill ranges. The tectonic boundary between internal and
external zone is marked by Nathiagali Thrust or Hazara Thrust, which is known as Panjal Thrust
(PT) (Baig and Lawrence, 1987; Fig. 2.3). The Precambrian basement and Paleozoic-Mesozoic
cover of the hinterland of the Indian plate lies north of the PT in Hazara, Kaghan and AJK. The
Paleozoic to Cenozoic folded and imbricated sedimentary sequence between the Nathiagali Thrust
and the SRT constitutes the foreland of the Himalayan collision zone.
The other important tectonic features of northwest Himalayas are HFT or Muzaffarabad Fault and
Jhelum Fault (Fig. 2.3). The Muzaffarabad Fault or HFT extends southeast through Jhelum valley,
Bagh and Poonch city in IAK areas (Baig and Lawrence, 1987; Baig, 2006). The Jhelum Fault is
north-south trending left lateral wrench fault, which separates the Potwar Basin from the Kashmir
Basin.
Chapter 2: Regional tectonics
14
2.3. Structural setting
The main structural features associated with tectonics of the Himalayas are described here briefly
as presented in Figs. 2.3 and 2.4.
2.3.1. Main Karakorum Thrust (MKT)
The MKT is also known as northern suture zone (Fig. 2.4; Tahirkheli, 1979). It marks tectonic
boundary between Kohistan Island Arc and Eurasian plate. The southern boundary of Karakorum
block, separates the Paleozoic metasediments of Karakorum block from the Cretaceous-Tertiary
Kohistan Ladakh Arc. The MKT was closed in late Cretaceous (Tahirkheli, 1982; Coward et al.,
1986).
2.3.2. Main Mantle Thrust (MMT)
The collisional boundary between the Kohistan Island Arc and the Indian plate is referred as MMT
(Fig. 2.4). The collision of Indian plate and Kohistan Island Arc occurred during Eocene time. It
extends west from Ladakh to northern Pakistan and eastern Afghanistan (Tahirkheli, 1979;
Gansser, 1981; Chaudhry et al., 1983). The MMT emplaces the lower crust crystalline rocks of the
Kohistan Island Arc on the Indian plate rocks (LeFort, 1975).
2.3.3. Main Central Thrust (MCT)
The MCT is the tectonic discontinuity along a major shear zone that separates the rocks of Higher
Himalayas and Lesser Himalayas (Heim and Gansser, 1939; Gansser, 1964; Fig. 2.4). The MCT in
AJK and Kaghan was proposed by Ghazanfar and Chaudhry (1986), Baig and Lawrence (1987)
and Chaudhry and Ghazanfar (1990). They reported a well developed 500 m to 5 km thick ductile
shear zone with inverse metamorphism and a sharp break in structural style and metamorphic
grade in Neelum valley, AJK and Kaghan valley, Pakistan. Part of MCT in Neelum valley AJK
proposed by Chaudhry and Ghazanfar (1990) was verified by Fontan and Schouppe (1995).
2.3.4. Panjal Thrust (PT)
The PT (Wadia, 1931; Calkins et al., 1975) defined one of the fundamental tectonic discontinuity
scars of the Lesser Himalayan domain in Kashmir (Fig. 2.4). The Pir Panjal range in Kashmir is
the type section of the thrust, from where it extends northeastwards along the eastern flank of the
Hazara Kashmir Syntaxis (HKS). The PT and MBT curves around the apex of the HKS then bend
southward. According to Wadia (1931), Calkins et al., (1975), Bossart et al., (1984) and Greco
(1989), these two faults join about 5 km north of Balakot (Figs. 2.3 and 2.4). The PT is parallel to
Chapter 2: Regional tectonics
15
the MBT on the eastern limb of the HKS but in the Hazara area below the PT, the MBT is not well
defined tectonic feature. It is an imbricated zone called as the MBT Zone (Baig and Lawrence,
1987).
Along the PT the low grade Precambrian Tanol Formation is thrusted onto unmetamorphosed
rocks of Cambrian to Jurassic age in the Hazara area. The Panjal Formation is exposed along the
Pir-Panjal range along the eastern flanks of HKS, whereas, it is not present below it in Hazara area
(Baig and Lawrence, 1987). According to Khan (1994), PT separates the rocks of Upper
Carbonferous to Triassic from Tanol Formation in Jhelum, Neelum and Kaghan valleys in the
northwest Himalayas.
2.3.5. Main Boundary Thrust (MBT)
The MBT is the main frontal thrust of the Himalayan range, which runs about 1500 km from
Assam in the east to Kashmir in the west. Wadia (1931) recognized a series of nearly parallel
faults in his division of outer Himalayas (also known as Sub Himalayas). All these were referred to
as the MBT. Presently, the outer most of the fault named Murree Thrust by Seeber and Armbruster
(1979) is called the MBT (Figs. 2.3 and 2.4). This distinct tectonic feature, in Pakistan has thrusted
the Eocene and older rocks over the Miocene Murree Formation. In the HKS and the fault loops
around HKS, the MBT is displaced by the left lateral active Jhelum Fault (Baig and Lawrence,
1987). This structure separates the rocks of the Lesser Himalaya (hanging wall) from the
sandstones, siltstones, clays and shales of the Sub Himalayas (footwall). The SW-directed
movement associated with this structure in Kashmir is characterized by brittle deformation
(cataclastites).
Chapter 2: Regional tectonics
16
Fig. 2.4 Enlarged regional tectonic map of northern Pakistan. Sources used as in Fig. 2.3. MKT:
Main Karakorum Thrust, MMT: Main Mantle Thrust, MCT: Main Central Thrust, PT: Panjal
Thrust, MBT: Main Boundary Thrust, HFT, Himalayan Frontal Thrust, JF: Jhelum Fault.
2.4. Indus Kohistan Seismic Zone (IKSZ)
A wedge-shaped northwest trending structure between MMT and HKS is known as IKSZ (Figs.
2.3 and 2.4; Armbruster et al., 1978; Seeber and Armbruster, 1979). Ni et al. (1991) confirmed the
presence of almost 100 km long feature between the HKS and MMT. This 50 km wide zone has
nearly horizontal upper surface and a northeast dipping lower surface of seismicity along IKSZ
(MonaLisa et al., 2008). The two seismic zones within the IKSZ have identified on the basis of
relocated hypocenters by Ni et al., (1991), a shallow zone extending from the surface to a depth of
8 km and a more pronounced mid crustal zone lying at depth of about 12 to 25 km. The upper
Chapter 2: Regional tectonics
17
boundary represents a decollement surface that decouples the sediment and metasediments from
the basement at a depth of about 12 km.
The IKSZ is seismically the most active structure capable of generating large earthquakes in the
region. It is predominantly a thrust fault with a northwest strike and northeast dipping plane
parallel to the general trend of the MBT to the northeast of Muzaffarabad (MonaLisa et al., 2008).
However, IKSZ with MBT is not comparable because both have different tectonic history, as
based on surface geology (Gahalaut, 2006). The most destructive earthquake prior to Kashmir
earthquake 2005 associated with IKSZ, was the 28 December, 1974 Pattan earthquake with
magnitude 6.0. The IKSZ represent the reactivation of decollement surface and have short term
stress field which may cause the broad zone of scattered seismicity and is responsible for the
Kashmir earthquake 2005 (MonaLisa et al., 2008).
2.5. Hazara Kashmir Syntaxis (HKS)
The Tertiary strata of the foreland basin and MBT take a hair-pin bend curvature around Hazara
and AJK in Pakistan from northwest to southeast and to south. This curvilinear pattern of
structural trend on regional scale was described as the NW Himalaya Syntaxis (Wadia, 1931)
which was later called as the HKS by various scientists (Calkins et al., 1975; Baig and Lawrence,
1987; Bossart et al., 1988; Greco, 1989; Kazmi and Jan, 1997, Rustam et al., 2003; Figs. 2.3 and
2.4).
The HKS is a NNW trending regional scale antiform structure that folds the Lesser- and Sub
Himalayas and to some extent also the Higher Himalayas. Except for the HFT, all the major thrust
including MBT, PT and MCT are refolded by this structure. Unlike the classical anticlines, where
the older rocks lie in the core and the limb comprises successively younger rocks, the HKS
exposes the youngest rocks in the core, and successively older rocks in the limbs. This owes to
crustal stacking due to thrusting prior to the development of the Hazara Kashmir Syntaxial
structure. Thus the limbs of the HKS comprise folded PT and MBT. The MBT, PT, Jhelum Fault
and Muzaffarabad Fault are the important active tectonic features in the HKS (Armbruster et al.,
1978; Baig and Lawrence, 1987; Baig, 2006; Yeats et al., 2006; Fig. 2.5).
2.6. Tectonic Models for the origin of HKS
• Wadia (1931) reported a SSE-ward and SW-ward transport direction on the western
and eastern limb of HKS respectively. He suggested an original horst to be involved in
the formation of HKS.
Chapter 2: Regional tectonics
18
• Calkins et al. (1975) agreed the Wadia (1931) model, but suggested that the west-
vergent shape of the anticlinal syntaxis results from a second, W-directed movement
direction.
• Bossart et al. (1984, 1988) described that the HKS formed in response to anticlockwise
rotation of the Indian plate during the late Tertiary. According to his model continued
rotation and contraction associated with the Indian plate resulted in development of the
kink structures into the crustal scale antiformal fold structures.
• Bossart et al. (1990) based on the paleomagnetic studies review Bossart et al., (1988)
model. Their finding suggested that where as the western limb did rotate anticlockwise
in Tertiary, however, the eastern limb rotated clockwise, opposite to the rotation of the
western limb.
• Treloar et al. (1992) proposed that HKS developed as a consequence of interference
between two active and converging thrust sheets. The Kashmir Himalayas in the east
and Hazara Potwar Himalayas in the west.
2.7. Core structures of the HKS
The HKS was formed due to the interaction of three independently moving tectonic elements;
the Himalayas, the Indo-Pakistan Shield and the Salt Range, each of which is moving
independently (Fig. 2.3). This may account for the unusual structure of the core of the syntaxis,
where major refolding patterns exist and over thrusting occurs on both sides of the syntaxis.
The HKS consists of a complex series of overlapping nappes made up of various Precambrian,
Palaeozoic and Mesozoic formations which have been over thrust on a group of predominantly
classic sediment, the Murree Formation of Tertiary age (Bossart et al., 1988). In the western
limb of HKS, the PT and MBT subdivided the region into three tectonic elements. The tectonic
element below the MBT in the core of the HKS is mainly composed of Tertiary sediments of
interbedded sandstone with shale and claystone of Murree Formation. Thus the Sub Himalayas
structurally the lowest element is characterized by intensive folding associated with the
formation of Muzaffarabad anticline. The core of anticline has carbonates of Cambrian,
Paleocene and Eocene. The tectonic element between MBT and PT mainly consists of Jurassic
to Eocene limestone, shales and Hazara slates of Precambrian age.
Chapter 2: Regional tectonics
19
Fig. 2.5 Structural map of the Hazara Kashmir Syntaxis (compiled and modified after Wadia,
1931; Calkins et al., 1975; Baig and Lawrence, 1987; Greco, 1991). HKS: Hazara Kashmir
Syntaxis; MBT: Main Boundary Thrust; PT: Panjal Thrust; HFT: Himalayan Frontal Thrust.
In the north of PT upper most tectonic units are formed by Tanol Formation of Precambrian age
and Cambrian Mansehra Granite (LeFort, 1981). Three structures in the core of the HKS are most
significant (Fig. 2.5).
i) Muzaffarabad Anticline
ii) Muzaffarabad Fault
iii) Jhelum Fault
2.7.1. Muzaffarabad Anticline
The north of Jhelum river near Muzaffarabad can be observed as an anticlinal structure, with a
well developed NE and SW limbs. The core of this structure is exposed near Muzaffarbad in the
north and hence the structure is appropriately named as the Muzaffarbad anticline (Fig. 2.5;
Chapter 2: Regional tectonics
20
Calkins et al., 1975; Hussain et al., 2004). This anticlinal structure is highly tectonized at its SW
limb due to its involvement in the active faults including PT, MBT, Jhelum Fault and
Muzaffarabad Fault marking the western limb of the HKS (Baig and Lawrence, 1987).
The anticline exposes the deepest stratigraphic levels in a 30 km long stretch between
Muzaffarabad and Balakot, where core is occupied by the Cambrian carbonates of Muzaffarabad
Formation. There is another set of carbonate rocks of Paleocene-Eocene age marked by
unconformable lower contact with the Cambrian Muzaffarabad Formation and transitional upper
contact with the Murree Formation. The Murree Formation occupies the entire eastern limb of
Muzaffarabad anticline as well as the northern apex of the HKS.
The western limb of Muzaffarabad anticline is highly diminishing because of involvement by the
faults marking the western limb of the HKS. This results in at places, complete absence of the
Paleocene-Eocene carbonate and shale units between the Muzaffarabad Formation and Murree
Formation. In another place, at this western limb of the syntaxis both the Paleocene-Eocene
carbonates as well as the Murree Formation are attenuated and the Muzaffarabad Formation is in
direct faulted contact with the Precambrian Hazara slates.
2.7.2. Muzaffarabad Fault (MF)
The Muzaffarabad Fault or HFT lie along the western limb of the HKS (Figs. 2.3, 2.4 and 2.5;
Baig and Lawrence, 1987; Nakata et al., 1991). The active Muzaffrabad Fault or HFT (Baig and
Lawrence, 1987) marks the western contact of the Muzaffarabad Formation and the Murree
Formation at the western limb of the Muzaffarabad anticline (Fig. 2.6). It is a thrust/reverse fault
exposed immediately east of Muzaffarabad. The fault is refolded along the Neelum river, before it
stretches NW towards the Balakot, where it crosses the PT and MBT at the apex of the HKS (Fig.
2.5; Baig and Lawrence, 1987; Baig, 2006). Nakata et al. (1991) showed an active fault along the
Jhelum river between Muzaffarabad and Grahi Dopatta, which he termed as Tanda Fault. The
Kashmir earthquake 2005 ruptured the Muzaffarabad Fault 120 km between Balakot in the north
to Bagh in the south (Avouac et al., 2006; Baig, 2006; Kanedo et al., 2008).
The Geological Survey of Pakistan (GSP) mapped a set of quadrangles that included northwest–
southeast-striking fault close to the IKSZ (Calkins et al., 2004; Hussain et al., 2004; Iqbal et al.,
2004; Hussain and Yeats, 2006) earlier mapped as HFT (Baig and Lawrence, 1987) that was not at
that time recognized as active but were later shown to be reactivated during the Kashmir
earthquake 2005 by Nakata and Kumahara (2006), Avouac et al. (2006), Baig (2006) and Kaneda
et al., (2008). Evidence for faulting includes juxtaposition of Cambrian Muzaffarabad Formation
against Murree Formation and Murree Formation against Kamlial Formation.
Chapter 2: Regional tectonics
21
Fig. 2.6 The Muzaffarabad Fault runs between the Muzaffarabad Formation and Murree Formation
in Makri area, Muzaffarabad city.
Immediately after the earthquake, the surface ruptures parallel to the reactivated Muzaffarabad
Fault from Balakot to Bagh areas were documented by Baig (2006) and Kaneda et al., (2008). The
field investigation has suggested that the typical geomorphology of the fault zone is a scarp with
compressional features at the base and tension cracks along the crest. Satellite imagery analysis
(JSCE, 2006) show that the maximum vertical uplift up to 5.5 m is along Muzaffarabad Fault. In
contrast, the field investigation indicates that the maximum uplift along fault is 7.5 m during
Kashmir earthquake 2005 (Baig, 2006). However, the maximum uplift along Muzaffarabad is 120
m which indicates multiple earthquakes uplift along fault in Late Holocene (Baig et al., 2008). The
mostly mass movements and ruptures are concentrated close to the fault on its hanging wall block.
2.7.3. Jhelum Fault (JF)
Jhelum Fault truncates the structure along the western limb of the HKS from Balakot to Kohala
(Baig and Lawrence, 1987; Fig. 2.5). The active evidence along the fault include offset of streams,
Chapter 2: Regional tectonics
22
tilted and deformed Quaternary terraces, nick points and dissected spurs (Baig and Lawrence,
1987). This is a left lateral strike slip reverse fault with offset of 31 km along the western limb of
the HKS (Baig and Lawrence, 1987). The fault is well exposed at Domel and Ambore areas in
Muzaffarabad. The Holocene conglomerates lies in the footwall block and the carbonates of
Precambrian Hazara Formation are in the hanging wall. The Jhelum Fault segment from Kohala to
Azad Pattan shows the Murree Formation in the hanging wall and Kamlial, Chingi and Nagri
Formations in the footwall (Baig et al., 2008).
Chapter 3: Geological setting
23
Geological setting
3.1. General description
In northern Pakistan, the Indian plate collided with the Kohistan Island Arc during the Tertiary
Himalayan collision. The basement and cover rocks of the Indian plate are deformed and
metamorphosed. The foreland fold-and-thrust belt of the Indian plate can be divided into three
main tectonic units. These tectonic units are the internal metamorphosed zone, external
unmetamorphosed to low grade metamorphosed zone and foreland basin sediments (Fig. 2.3).
The internal metamorphosed zone is between the PT and MMT. This zone comprises of
metamorphosed cover and basement rocks of the Indian plate. The basement rocks are
Precambrian metasediments and gneisses intruded by lower Paleozoic granites (Greco, 1989). The
cover rocks and basement are metamorphosed from greenschist to amphibolite facies. The external
unmetamorphosed to low grade metamorphosed zone are between the PT and MBT. The rocks
include Precambrian to Tertiary cover sequence of Indian plate (Latif, 1970; Calkins et al., 1975).
The external zone is imbricated and folded. The foreland basin sediments are bounded by the MBT
and SRT. The SRT delineates the southern most extent of the foreland basin sediments. The
Eocambrian to Tertiary sedimentary cover rocks are exposed in the basin. These Himalayan
tectonic units are folded to form the HKS (Wadia, 1931; Calkins et al., 1975; Bossart et al., 1988).
3.2. Geology of Hazara Kashmir Syntaxis
The most prominent geological structure in the western Himalayas in northern Pakistan is known
as HKS. It covers the most complex region of the mountain range. The sedimentary to
metamorphic rock sequence is imbricated and folded to form HKS during the Tertiary Himalayan
collision (Wadia, 1931; Calkins et al., 1975; Baig and Lawrence, 1987; Bossart et al., 1988;
Greco, 1991). The PT at the apex and eastern limb of HKS marks tectonic boundary between the
Carboniferous-Triassic Panjal Formation and Precambrian metamorphosed Tanol Formation. The
Tanol Formation is intruded by Cambrian Mansehra granite. Whereas, along the western limb of
the HKS, PT separates the Precambrian Tanol Formation from the Precambrian Hazara Formation,
Cambrian Abbotabad Group, Jurassic-Cretaceous rocks and Paleocene-Eocene sequence (Fig. 3.1).
The geology in the core of the HKS includes carbonates of Cambrian Muzaffarabad Formation,
Paleocene- Eocene limestones, shale, clay, siltstone and sandstone sequence and Miocene Murree
and Kamlial Himalayan molasse (Calkins et al., 1975; Bossart et al., 1988; Greco, 1989; Munir et
al., 2006; Baig and Snee, 1995; Fig. 3.1).
Chapter 3: Geological setting
24
These sediments lie in the foot wall block of the MBT. The MBT and PT wrap around the Murree
Formation which lies in the core of the HKS. The thrust zone between PT and MBT is narrow
along the eastern limb and wider along the western limb of HKS (Baig and Lawrence, 1987).
Fig. 3.1 Map showing the geological units and faults of Hazara Kashmir Syntaxis (compiled after
Wadia, 1931; Latif, 1970; Calkins et al., 1975; Baig and Lawrence, 1987; Greco, 1989; Hussain et
al., 2004; Munir et al., 2006 and Kaneda et al., 2008).
Chapter 3: Geological setting
25
3.3. Geological setting of the study area
The regional stratigraphy of HKS shows sedimentary, metasedimentary, metavolcanics and
metaigneous rock units. These lithostratigraphic units have different geological control on the
mass movements of the area. The Precambrian to Tertiary rocks are exposed in the HKS around
Muzaffarabad areas (Fig. 3.2). The rock sequence includes the Precambrian Hazara and Tanol
Formations, the Cambrian Muzaffarabad Formation and Mansehra Granite, the Carboniferous-
Triassic Panjal Formation, the Paleocene-Eocene sequence, the Early Miocene Murree Formation,
the Late Miocene Kamlial Formation and Quaternary sediments.
Fig. 3.2 Geological map of Jhelum valley, Neelum valley and Muzaffarabad city (compiled and
modified after Wadia, 1931; Latif, 1970; Calkins et al., 1975; Baig and Lawrence, 1987; Greco,
1989; Hussain et al., 2004, Munir et al., 2006 and Kaneda et al., 2008).
Chapter 3: Geological setting
26
The earlier workers like Wadia (1931), Calkins et al., (1975), Greco (1991), Baig and Snee (1995),
Hussain et al., (2004), Khan (1994), Iqbal et al., (2004) and Munir et al., (2006) worked on the
stratigraphy and structure of the area. The stratigraphic sequence affected by Kashmir earthquake
2005 in Muzaffarabad and surrounding areas is presented in Table 3.1.
Table 3.1 Geology of the earthquake affected area.
Name Lithology Age
Quaternary Stream bed deposits and alluvium.
Holocene
Kamlial Formation Sandstones, shales, claystones and minor intraformational conglomerates.
Late Miocene
Murree Formation
Interbedded sandstones, siltstones with shales
and claystones.
Early Miocene
Paleocene-Eocene (Hangu, Lochart, Patala, Margala,
Chorgali and Kuldana Formations)
Nodular limestones, calcareous and
carbonaceous shales, claystones and laterite.
Paleocene-
Eocene
Panjal Formation
Metacarbonates, metasediments, metabasic volcanics, quartzite and graphitic schists.
Carboniferous-
Triassic
Muzaffarabad
Formation
Manshera Granite
Cherty and stromatolitic dolomites, cherty white
and grey bands, limestones and black shales
Coarse grained two-mica granite gneiss.
Cambrian
Cambrian
Tanol Formation
Pelitic and psammitic metasediments,
subordinate minor graphitic schist, talc schist and marbles.
Precambrian
Hazara Formation
Slate, phyllite and shales with minor limestones
and graphitic layers.
Precambrian
The brief description of rock units involved in Kashmir earthquake 2005 within study area is as
follows.
3.3.1. Hazara Formation
The Precambrian Hazara Formation is well exposed in AJK along the western limb of HKS (Figs.
3.1 and 3.2). In Muzaffarabad, the formation is exposed on Muzaffarabad-Mansehra road and
along the right bank of Jhelum river (Fig. 3.3). The lower contact of Hazara Formation with the
Chapter 3: Geological setting
27
Murree Formation is faulted, whereas the overlying contact with the Abbotabad Formation is
unconformable in Abbotabad area (Latif, 1974).
Fig. 3.3 The slates of Hazara Formation are well exposed along Muzaffarabad–Mansehra road in
Muzaffarabad area.
Lithologicaly, the formation consists of slates, phyllite, metasandstone and shale with minor
limestone and graphitic layers. The limestone with quartzite layers occur within the Hazara
Formation along the MBT from Domel Muzaffarabad to Rara and extended further southward
(Khan, 1994). The fresh color of slate and phyllite is black and green to dark green, however, the
weathering color is brown and dark green. The color of limestone is dark grey to greyish, whereas,
the quartzite is greenish to whitish grey (Calkins et al., 1975). The limestone beds with maximum
thickness of 150 meters and a sequence of calcareous phyllite and gypsum ranging from 30 to 120
meters thick are found in southern most Hazara and AJK (Calkins et al., 1975). The sedimentary
structures like fine lamination graded bedding and cross bedding can be seen in the slates.
Crawford and Davis (1975) analyzed three samples of low grade fine grained clastic rocks from
the Hazara Formation for age determination by the Rb-Sr method and obtained ages of 765±20 and
950±20 Ma. Calkins et al. (1975) assigned Precambrian age to the Hazara Formation.
Chapter 3: Geological setting
28
3.3.2. Tanol Formation
The Tanol Formation includes pelitic and psammitic metasedimentary rocks with local subordinate
graphitic schist, talc schist and marbles (Table 3.1). It is intruded by doloritic and granitic
intrusions. The unit is well-exposed in the south and southeastern margin of the Mansehra Granite
and around the HKS (Fig. 3.2). The Tanol Formation mainly consists of medium grained quartzite
and fine grained mica-quartz schist south of Mansehra Granite. The grade of metamorphism in
Tanol Formation increases from south to north.
The thickness of the Tanol Formation is difficult to measure due to structural complications.
Marks and Ali (1962) estimated the thickness ~ 1,600 m. At quite a few places, the Tanol
Formation is missing and the Hazara Formation underlies the Abbottabad Formation. The Tanol
Formation underlies the Abbottabad Formation and overlies the Hazara Formation in the area
between Abbottabad and Indus river. The contact between the Abbottabad Formation and the
Tanol Formation in this area is marked by an unconformity which is represented by a boulder bed
known as Tanakki conglomerate.
The Tanol Formation is devoid of fossils. However, from the above mentioned contact relation it is
evident that the Tanol Formation is younger than the Hazara Formation of Precambrian age and
older than the Abbottabad Formation. Latif (1974) assigned the age of the Tanol Formation is
Precambrian on the basis of Cambrian fossils in the Abbottabad Formation.
3.3.3. Muzaffarabad Formation
The cherty and stromatolitic dolomites, cherty white and grey bands, limestones and black shales
are called Muzaffarabad Formation (Baig and Snee, 1995; Figs. 3.1 and 3.2). The Muzaffarabad
Formation is well exposed in northeast of Muzaffarabad city (Fig. 3.4) in the core of Muzaffarabad
anticline. The fresh color of Muzaffarabad Formation is grey to dark grey and weathered color is
light brown and grey. The rock unit has sedimentary breccia and conglomerate layers on the top.
Stromatolite are developed in the dolomite and its good exposures are found at the Neelum valley
section. The dolomites are thin to thick bedded. The estimated thickness of Muzaffarabad
Formation is ~ 800 meters in the northeast of Muzaffarabad (Calkins et al., 1975).
The lower contact of Muzaffarabad Formation is faulted with Murree Formation along
Muzaffarabad Fault, while the upper contact with Paleocene-Eocene sequence is unconformable.
The Muzaffarabad Formation along this fault is highly crushed and sheared for a width of about
150-300 meters to the north and northeast of Muzaffarabad city. The Muzaffarabad Formation is
stratigraphic equivalent to fossiliferious Cambrian Abbottabad Formation of Latif (1974). The age
of Muzaffarabad Formation is Cambrian.
Chapter 3: Geological setting
29
Fig. 3.4 The highly crushed and sheared bedded Muzaffarabad Formation exposed at Kamsar area,
in the northeast of Muzaffarabad city.
3.3.4. Mansehra Granite
The coarse grained two-mica granite gneiss intrudes Tanol Formation. It is known as Mansehra
Granite. The rock unit is well exposed within the tectonic unit of Tanol Formation in the eastern
and western limbs of HKS (Greco, 1989; Figs. 3.1 and 3.2). These gneisses are coarse grained with
augen structure. The Mansehra Granite is composed of quartz, feldspar, muscovite, biotite, epidote
and tourmaline minerals. The rock unit has been assigned Cambrian age (LeFort et al., 1980).
3.3.5. Panjal Formation
The formation is well exposed in Neelum and Kaghan valleys. Midlemiss (1910) mapped the belt
of metacarbonates, metasediments, metavolcanics, quartzite and graphitic schists at the apex of the
HKS. Wadia (1931) called it Panjal Volcanics Series and assigned the name of Panjal Formation.
In Muzaffarabad area, the Panjal Formation lies between the MBT and PT (Figs. 3.1 and 3.2).
The Panjal Formation consists of two units namely Panjal metasediments and Panjal volcanics.
Lithologically, the formation consists of metacarbonates, metasediments, metavolcanics, quartzite
Chapter 3: Geological setting
30
and graphitic schists (Calkins et al., 1975). The carbonate rocks of the Panjal Formation are
crystalline limestones and dolomitic limestones. The Panjal volcanics consist of green to greenish
grey basaltic lava flows with tuffaceous layer and subordinate intercalation of limestone. The
volcanics are characterized by massive lava flows and pillow lava with intercalations of limestone
and bedded chert. Sequence of this formation at some places also contains serictic, quartzitic and
metapelitic rocks with subordinate lenses of volcanics. According to Calkins et al., (1975), the
volcanic greenstone generally display a weakly developed schistosity in most places parallel to the
original layering. The volcanics were metamorphosed to lower greenschist facies during
Himalayan orogeny. Fossils have been discovered from various localities which indicate Triassic
age. The Panjal Formation has been assigned an Upper Carboniferous-Triassic age (Wadia, 1928
and 1931).
3.3.6. Paleocene-Eocene sequence
In this study, the Hangu, Lockhart, Patala, Margala, Chorgali and Kuldana Formations are
compiled collectively as Paleocene-Eocene sequence (Figs. 3.1 and 3.2).
3.3.6.1. Hangu Formation
The rocks of Hangu Formation in Muzaffarabad are exposed along the Neelum valley road section
at Yadgar, Batmang and towards east at Khilla, Maira Tanolian and Tanda Botha areas. The
formation is thinly developed and occurs in patches along Muzaffarabad Formation.
Lithologically, the formation consists of brecciated quarzite, bauxite, limonite, fire clay,
carbonaceous shales, sandstones, coal seams and conglomerates (Ashraf et al., 1989). The shales
of Hangu Formation are grey and sandstone is light grey and reddish brown. The sandstone is fine
to coarse grained and medium bedded. The limonite and bauxite are found near Yadgar area of
Muzaffarabad. The lower contact with Muzaffarabad Formation is unconformable, whereas the
upper contact with Lockhart Formation is sharp. The age assigned to Hangu Formation is Early
Paleocene (Munir and Baig, 2006; Munir et al., 2006).
3.3.6.2. Lockhart Formation
The Lockhart Formation is exposed in Muzaffarabad near Yadgar, Khilla and Tanda-Botha areas.
The formation consists of limestones and subordinate shales. These limestones are thinly to
medium bedded with marly intercalations. At places it is nodular and nodularity increases toward
the top of formation. The nodules are generally 2-6 cm in length and 1-5 cm in width (Munir et al.,
2006). The limestone is grey to dark grey on fresh surface, while dirty grey to light grey on
Chapter 3: Geological setting
31
weathered surface. The lower and upper contacts are conformable with Hangu Formation and
Patala Formation respectively. The age assigned to this formation on the basis of fossils is Early
Paleocene (Munir and Baig, 2006).
3.3.6.3. Patala Formation
The outcrop of Patala Formation is well exposed near Yadgar and Tanda-Botha areas of
Muzaffarabad. The Patala Formation consists of shales, claystone, siltstone and sandy limestone.
The shales are dark to greenish grey. These are carbonaceous and calcareous. The limestone is
white to light grey and nodular. These shales contain coal seams. The lower contact of Patala
Formation with Lockhart Formation is gradational, while the upper contact with Margala Hill
Formation is transitional. The formation is richly fossiliferious and contains abundant foraminifers.
On the basis of fossils, the formation is Late Paleocene (Munir and Baig, 2006; Munir et al.,
2006).
3.3.6.4. Margala Hill Formation
The Margala Hill Formation is well exposed in Muzaffarabad along Neelum valley road section.
Lithologically, the formation consists of limestone with subordinate marl and shale. It is hard,
massive and thin to thick bedded at Yadgar near Muzaffarabad (Fig. 3.5). The limestone is grey to
dark grey on fresh broken surfaces, whereas the weathered color is pale grey. The limestone is
nodular, fine to medium grained and hard. The marl is grey to brownish grey, while the shale is
greenish brown to brown.
The lower contact of the Margala Hill Formation with Patala Formation is gradational in Kaghan
area (Ghazanfer and Chaudhry, 1986), whereas the upper contact with Kuldana Formation is
transitional in Muzaffarabad area. Munir and Baig (2006) recorded a number of foraminifers from
the formation and assigned an Early Eocene age.
Chapter 3: Geological setting
32
Fig. 3.5 Margala Hill limestone exposed at Yadgar area, Muzaffarabad.
3.3.6.5. Chorgali Formation
The Chorgali Formation is exposed in Yadgar, Khilla and Tanda-Botha areas in Muzaffarabad.
Lithologically, the formation consists of shales, limestone and dolomitic limestone (Munir et al.,
2006). The shales are calcareous and grey to greenish grey. The dolomitic limestone is white to
light grey and platy.
The lower and upper contacts of the Chorgali Formation with Margala Hill Formation and
Kuldana Formation are gradational. On the basis of fossils, the age assigned to Chorgali Formation
is Early Eocene (Munir and Baig, 2006).
3.3.6.6. Kuldana Formation
In Muzaffarabad, the rocks of Kuldana Formation are exposed along the Neelum valley road
section. In this area, the Chorgali Formation passes gradually into the Kuldana Formation.
Lithologically, the Kuldana Formation consists of maroon to dark red clays and shales with
subordinate green to greenish grey shales and fine grained sandstone. The lower contact with the
Chorgali Formation is gradational, while upper contact with the Murree Formation is transitional.
On the basis of fossils Munir et al., (2006) and Munir and Baig (2006) assigned early to middle
Eocene age to the Kuldana Formation.
Chapter 3: Geological setting
33
3.3.7. Murree Formation
The Murree Formation occupies the major extent of the study area from south to north in Jhelum
and Neelum valleys (Figs. 3.1 and 3.2). It has faulted contact with Panjal Formation along MBT in
the eastern limb of the HKS. It is well exposed in Chalpani, Ghori, Pattika and Panjgran areas of
Neelum valley. In Jhelum valley, it is exposed near Niazpura, Kardala, Langarpura, Paprusa, Khun
Bandi and Hattian Bala areas. It is also well exposed along Muzaffarabad to Kohala road (Fig.
3.6).
The Murree Formation consists of interbedded sandstones, siltstones, shales and claystones. The
sandstone displays cross bedding and ripple marks. The sandstone is medium grained and medium
to thick bedded. The weathered color of sandstone is grey to dark grey and reddish brown while
fresh color is grey. The calcite and quartz veins are abundantly present within the sandstone. The
siltstone is thin bedded. The shales of Murree Formation are reddish brown.
Fig. 3.6 Fractured and jointed sandstone of the Miocene Murree Formation exposed in the Jhelum
valley area.
Wadia (1928) divided Murree Formation into lower and upper parts on lithological basis. These
two parts have been identified in the field in Jhelum valley. Generally the sandstone of Lower
Chapter 3: Geological setting
34
Murree Formation is hard and fine grained and is associated with red purple colored shales. In the
upper Murree Formation, the sandstone is soft, coarse grained and micaceous with pale grey to
brownish grey in color. The beds of sandstone, clay and shale alternate with each other showing
that the Murree Formation is the result of cyclic deposition. The Murree Formation represents the
molasse of the Himalayan orogeny.
The Murree Formation shows a considerable thickness in AJK. Calkins et al. (1975) have recorded
a thickness of 1500 meters of the Murree Formation in Muzaffarabad area. The lower and upper
contacts of Murree Formation with Kuldana Formation and Kamlial Formation are transitional
respectively. The age assigned to the Murree Formation is Early Miocene (Munir and Baig, 2006).
3.3.8. Kamlial Formation
The Formation is exposed along the left bank of the river Jhelum in Kumar Bandi and Garhi
Dopatta areas (Fig. 3.2). The formation mainly consists of sandstones, shales, claystones and
minor intraformational conglomerates. The sandstone is purple grey and blackish grey, medium to
coarse grained, hard and compact. The Kamlial Formation is differentiated from Murree
Formation by its spheroidal weathering and dominance of mineral tourmaline over epidote. The
formation overlies the Murree Formation with a transitional contact while the upper contact is not
exposed in the study area. The age of the Kamlial Formation is Late Miocene.
3.3.9. Quaternary Sediments
The Quaternary sediments include alluvium and colluvium deposits are well exposed within the
study area and are briefly described as follows.
3.3.9.1. Alluvium deposit
The alluvium deposits consist of boulders, cobbles, gravel, sand, or silt. The deposits in the
Muzaffarabad city show the braided stream deposition. These deposits unconformably overlie the
bedrocks of different ages. The alluvium deposits are present throughout the area. In Chellah
Bandi, the alluvium unconformably overlies the bedrock of Murree Formation. At the base of
alluvium, the braided stream deposits of Neelum river that include boulders, cobbles and pebbles
of the Tanol Formation, Panjal Formation, granites, gneisses, marbles, sandstone, dolomite and
quartzite in sandy and silty matrix. The braided stream deposits at places have the layers and
lenses of sand, silt and clay are present. Generally, on the top of the braided stream deposits there
are deposits of 2-3 meters layer of sand, silt and clay.
Chapter 3: Geological setting
35
The alluvial fans are present along the left bank of Neelum river in Chellah Bandi, Dhanni, Plate,
Dherian and Maira Tanolian areas (Fig. 3.7). These alluvial fans mostly include the angular to sub-
angular fragments of Muzaffarabad Formation, whereas fragments of Murree Formation are mixed
at the base. At places, these alluvial fans are intermingled with the river alluvium.
The alluvial deposits present in Gojra and Chatter Domel areas unconformably overlie the Hazara
Formation. The alluvial fans in these areas mostly include the fragments of Hazara Formation. The
terraces present in Maira Kalan, Miani Bandi, Langarpura, Paprusa and Khund Bandi in Jhelum
valley areas are the alluvium deposits of Jhelum river. These deposits include boulders, cobbles
and pebbles of Panjal Formation, Siwaliks, quartzites and sandstones embedded in sand, silt and
clay matrix. The alluvial fans in these areas include angular to sub-angular fragments of sandstone
and siltstone of Murree Formation in clayey and silty matrix.
3.3.9.2. Colluvium deposit
The material that accumulates at the base of the slope as the result of the gravity or any other
triggering factor is called colluvium. It includes rock fall deposits that accumulate at the base of
the talus. Similarly, landslide, slump and debris deposits formed from any of the surficial materials
as defined above.
The colluvium deposits are the angular to sub-angular fragments of Muzaffarabad, Murree and
Hazara formations in Muzaffarbad city. In Muzaffarabad, these deposits are present in Makri Nala,
Shahwi Nala and upper Rinjata areas (Fig. 3.8). These deposit lies mainly at the base of the steep
cliffs along the streams. The colluvium deposits present in Neelum valley belongs to the Murree
Formation. However, at Nausari these deposits belong to the Tanol Formation. Likewise, in
Jhelum valley area, these deposits are present along the Jhelum river and belong to the Murree
Formation.
Fig. 3.7 The alluvial fan deposits near
Chellah Bandi, Muzaffarabad.
Fig. 3.8 Talus slope deposit at the base of
cliff at Shahwi nala, Muzaffarabad
Chapter 4: Methodology
36
Methodology
4.1. Introduction
The methodology used to investigate the mass movements triggered by the Kashmir earthquake
2005 is similar to those described in the existing literature. Medium and high resolution satellite
data (SPOT-5 (2.5 m), Quickbird (0.6 m)), Digital Elevation Models (ASTER and SRTM), field
data and published literature were used in this study. Satellite imageries supplemented with field
data were used for the statistical analysis of the mass movement distribution. The principal
methodological steps followed during the research work are presented in Fig. 4.1. A detail account
of methodology for mass movement distribution analysis and empirical analysis has been
discussed under following sections.
4.2. Available resources
4.2.1. Literature collection
The existing scientific literature relevant to the mass movement investigation included published
research papers, technical reports and maps covering the field of geology, seismology and
geotechnical investigation were collected. They were mostly produced after the Kashmir
earthquake 2005. However, unpublished investigations before the earthquake, though with
sporadic information and unsystematically recorded, have also provided some information on the
mass movement events. The unpublished technical reports and theses work carried out after the
earthquake by Japan International Corporation Agency (JICA), Japan Society of Civil Engineers
(JSCE), National Engineering Services Pakistan (NESPAK), Earthquake Reconstruction and
Rehabilitation Authority (ERRA), GSP, Azad Kashmir Mineral and Industrial Development
Corporation (AKMIDC) and Institute of Geology, University of Azad Jammu and Kashmir
(UAJK) were also consulted. This information generally deals with the mass movements and their
impacts, however, some basic information about mass movement events triggered by the Kashmir
earthquake 2005 was also provided in some of the literatures.
4.2.2. Topographical maps
Topographical maps at a scale of 1: 50,000 with contour intervals of 30 meters from the Survey of
Pakistan (Sheet No. 43F/7, 43F/11, 43F/12 and 43F/16) were used in this study. These maps were
obtained from the Planning and Development department AJK, Muzaffarbad. They were scanned
and geo-referenced by using the projection parameters given in the map. These maps being very
old and often generalized with low accuracy and poor representation of contour lines and relief
Chapter 4: Methodology
37
were used as the base map of the area, for the identification of locations and the topographical
features.
Fig. 4.1 Methodological steps used during the research work.
Literature
review
Geol. data
DEMs Satellite imagery
Data analysis
Classification Distribution map Case
studies
Database inventory
Statistical
analysis
Empirical
analysis
Conclusions
Start
Available
data
Field
survey
Photographs
Profiles
Mapping
Chapter 4: Methodology
38
4.2.3. Geological maps
A geological map of the investigated area prepared in April 2004 by the GSP was digitized and
modified at a scale of 1:50,000. A number of previously published maps (Wadia, 1931; Calkins et
al., 1975; Baig and Lawrence, 1987; Greco, 1991; Hussain et al., 2004; Kaneda et al., 2008) were
used to compile and refine the present geological map of the study area. A geological map of
Hattian Bala area prepared by GSP after the Kashmir earthquake 2005 was digitized and modified
at a scale of 1: 10,000.
4.2.4. Satellite imagery
The processed satellite imageries (SPOT-5 and Quickbird) of the study area, taken immediately
after the Kashmir earthquake 2005 were obtained from the Bundesanstalt für Geowissenschaften
und Rohstoffe (BGR) and GSP. These satellite imageries were used for the identification of the
mass movement events triggered by the earthquake, prior to the field investigation. Visual
interpretations of mass movements were made simply by identifying the areas where vegetation
was removed due to the downslope movement.
4.2.5. Digital Elevation Models (DEMs)
The pre- and post-earthquake topographic contour lines were derived from the ASTER and SRTM
based Digital Elevation Models (DEMs). They were used during field investigation. An ASTER
based DEM was used to calculate the topographic parameters (slope angle, slope aspect, and
elevation) for the statistical analysis of the mass movement distribution triggered by the
earthquake.
4.3. Field survey
Field surveys were conducted during 2009 and 2010 to collect the field data for the investigation
of the mass movements. Multiple field visits were carried out (March-April, 2009; October-
November, 2009; September-October, 2010) to collect the field data systematically. Prior to the
field visits, existing scientific literature was thoroughly reviewed. In addition, topographic,
tectonic, seismological, geological, and mass movement distribution maps were collected and
examined in details.
The Kashmir earthquake induced mass movements are clearly visible on the medium and highly
resolution SPOT-5 (2.5 m) and Quickbird (0.6 m) imageries (Fig. 4.2). The SPOT-5 images cover
the whole study area, while the Quickbird imageries were only available for an area of about 271
Chapter 4: Methodology
39
km2 of Muzaffarbad city and its surroundings. From these satellite images by visually inspecting,
the locations and extent of mass movements were easily mapped.
Fig. 4.2 SPOT-5 (A) and Quickbird (B) imagery show the mass movements triggered by the
Kashmir earthquake 2005 in Botha area Muzaffarbad.
Using these satellite imageries, 1,460 mass movement events were interpreted within an area of
approximately 1,299 km2 in the vicinity of Muzaffarabad city, Jhelum valley and Neelum valley
areas (Fig. 4.3). Of these, 127 mass movements were identified for field investigation. Four case
studies (Hattian Bala rock avalanche, Langarpura and Neelidandi rock falls, Panjgran slump and
rock fall) were selected for detailed geotechnical mapping. These large scale mass movements
were geologically and structurally mapped at a scale of 1:10,000.
The mass movement events identified from the satellite imageries for field investigation were
visited, observed and cross checked during the field work in October 2009 and September 2010.
The criteria used for the identification and investigation of mass movements in field included the
visits of the general aspect of the slope failure, area above detachment zone, detachment zone,
transport area and deposit area. The general aspect included the triggering causes, damages, and
victims. However, all mass movements triggered by the Kashmir earthquake 2005 were identified.
A B
Chapter 4: Methodology
40
Fig. 4.3 The mass movements interpreted from SPOT satellite imageries and field investigation
within the study area.
The mass movements triggered before the earthquake were almost reactivated during the Kashmir
earthquake 2005. This was verified during the field visits and interviewing the local residents on
sites. The damages and victims were assessed only for large scale mass movements. The data
collection of the area above the detachment zone included: altitude, average slope angle, geometry,
presence of cracks, geology, soil cover, weathering phenomena and hydrology. The data collection
of the area of the detachment zone covers: geometry, geology, morphology, rock type, and
process. Transport and deposit area were used as criteria for mass movement travel distance and
Chapter 4: Methodology
41
volume of the deposit. This includes the geometry, length profile, Fahrböschung angle, shadow
angle, talus slope angle, deposit and material (Appendix 1).
During the following field visits, systematic measurements were taken by using the Braunton
Compass, Laser distance meter, Global Positioning System (GPS), Clinometer and measuring tape.
Field methods were adopted following Willianson et al., (1991). Braunton Compass was used to
measure the attitude of the bed rock and dip of the scarp. The elevation and geographic coordinates
of mass movement were taken using portable GPS receiver (Garmin eTrex series, with an accuracy
of ± 10 meters). GPS waypoints were gathered at specific locations around the mass movements to
map the scarp and the body of the mass movements in the field. However, these data were not
collected for all mass movement events identified in the field due to inaccessible high mountain
terrain and steep slopes. As a result, GPS way points were gathered only for larger scale mass
movements or wherever it was possible. A laser distance meter (RIEGL FG21-HA, with an
accuracy of ± 1meter) was used for absolute horizontal measurement. These field measurements
were used to construct the geological longitudinal profiles. Clinometer was used to measure the
Fahrböschung angle, shadow angle and talus slope angle on site in field. The length and width
were measured directly on site during field visits. Measuring tape was used to measure the
dimension of the mass movements. The depth was estimated by judging the height of the shear
surface or the thickness of the deposit material. The volumes of mass movements were estimated
by multiplying the deposit area by an estimated average thickness.
The topographic maps were enlarged at a scale of 1:10,000 after they had been geo-referenced.
These maps supplemented with satellite imageries were used to map the mass movement events
during the field visits. Field mapping included scarp, body and deposit of the mass movement. The
tensional cracks and any other specific features (secondary scarp, morphology etc.) were also
included. For detailed mapping of large scale mass movements, the geology and structural features
were also mapped. Physical features of the mass movements were observed and noted during the
field visits. Many features were examined to get the information on the size and characteristics of
the mass movements. However, inaccessible physical features were viewed from the road or at the
base of mass movements and photographed for further description of the features of these mass
movements.
Chapter 4: Methodology
42
Fig. 4.4 Examples of the mass movements triggered by the Kashmir earthquake 2005 in NW
Himalayan, Pakistan. A. Subri rock fall blocked the Jhelum valley road during the earthquake, B.
Rockslide of Saran area caused to damage the houses, C. Dehrian Saidan rock fall on hills of
Muzaffarabad city, D. Debris fall near the Pattika town, E. Author measuring absolute horizontal
distance of mass movement in field, and F. Fractured sandstone in Miocene Murree Formation at
Makri, Muzaffarabad.
Chapter 4: Methodology
43
4.4. Database inventory
For any type of mass movement investigation a correct inventory database is pre-requisite (Varnes,
1984). Unfortunately before the earthquake 2005 the mass movement information was sporadic
and no systematic investigation was carried out. The scientific papers and technical reports for the
mass movement investigations about the study area were produced by international and national
scientific communities after the Kashmir earthquake 2005. Engineering Cell, ERRA first time after
the earthquake collected the information of 64 mass movements along the road side in the Jhelum
valley and the Neelum valley area. This information was of initial level and relatively incomplete
to be used for further scientific investigations. Owen et al. (2008) developed an inventory data
base containing 1,293 mass movements at 174 locations. This data base contains only information
about locations and types of failures. Therefore, it was necessary to develop a complete database
including all information such as length, width, height, depth, volume, area, Fahrböschung angle,
shadow angle and talus slope angle as a pre-requisite to investigate the mass movements triggered
by the Kashmir earthquake 2005. The following available resources and methods were used to
prepare the mass movement inventory data base for this study:
• Published research papers and technical reports
• Satellite imageries
• Extensive field visits using field mapping
• GPS survey
• Questionnaire / record sheets
The characteristics of all mass movement events mapped during the field investigations were
examined and verified with the photographs taken and the questionnaires filled during the field
work. The data base of 127 mass movement events triggered by the Kashmir earthquake 2005 was
prepared. The database inventory included: geographic coordinates, locations, type, elevation at
top, elevation at toe, length, width, height, depth, Fahrböschung angle, shadow angle, talus slope
angle, surface area, deposit area, volume, triggering factor, nature of material, predominant bed
rock, boulder size, cracks, vegetation and geological formations. The surface area and the deposit
area were calculated for all mapped mass movements after digitizing by using ArcGIS 9.3
software (Fig. 4.5).
Chapter 4: Methodology
44
Fig. 4.5 Mass movements triggered by the Kashmir earthquake 2005 data base interface prepared
in ArcGIS 9.3.
The data for statistical analysis of mass movement distribution was prepared using the satellite
imageries incorporating field data and ASTER based DEM. Numbers, distances, mass movement
concentration and areas of mass movements triggered by the earthquake were calculated. The
methodology adopted for this analysis is described in detail in chapter 6 (see for detail in section
6.2)
For the empirical analysis, 103 mass movement events were selected (Appendix II). These include
the parameters: length, height, depth, surface area, deposit area, volume, Fahrböschung angle,
shadow angle and talus slope angle. These mass movement events were classified as: rock fall,
debris fall and mountain fall based on the volumemetric classifications of Varnes (1978) and Fell
(1994). The detail description of methodology used for empirical analysis in this study is described
in chapter 7 (see for detail in section 7.3).
4.5. Data analysis
The mass movement classification map was prepared with data collected by own field
investigations. The mass movements were classified as: landslide, rock fall, debris fall, rock
avalanche, slump and rock fall during field observations according to the classification system of
Varnes (1978). The geographical distribution maps of Muzaffarabad city, Jhelum valley and
Chapter 4: Methodology
45
Neelum valley areas were prepared based on field data and satellite imageries. These maps were
produced in Arc Map of ESRI ArcGIS 9.3 software. The maps were georeferenced to the
Universal Transverse Mercator (UTM) projection system and oriented to the 43 North datum WGS
(1984).
The mass movement distribution map of the study area was prepared at a scale of 1:50,000. This
map is the base for the statistical analysis of the mass movement distribution. The 1,460 mass
movements were interpreted using points for initiation zones and polygons for the surface area.
ArcGIS 9.3 was used to analyze the relationship among the mass movement distribution and
causal factors. Correlation and linear regression methods were used to determine the relationship
between the distance from the epicenter and active Muzaffarabad Fault and the mass movement
distribution. Empirical models were adopted for the empirical analysis to determine the
relationships between the various geometrical parameters of mass movement events collected in
the field.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
46
Mass movements triggered by the Kashmir earthquake 2005
5.1. General overview
Mass movements are generally associated with a trigger such as an earthquake, heavy rain fall and
human activity. An earthquake having a minimum magnitude of 4.0 might cause mass movements
from vulnerable slopes, and an earthquakes with higher magnitudes can cause increased numbers
of mass movements, covering an area of about 10,000 to 40,000 km2 (Keefer, 1984, 1999, 2002).
The Kashmir earthquake 2005 having magnitude 7.6, triggered thousands of mass movements
throughout an area of more than 7,500 km2 (Fujiwara et al., 2006; Sato et al., 2007; Kamp et al.,
2008; Owen et al., 2008). However, this earthquake induced less numbers of mass movements in
the area as compared to other earthquakes in the world (Harp and Jibson, 1996; Hung, 2000;
Keefer, 2002; Khazai and Sitar, 2003; Qi et al., 2010).
The area affected by the Kashmir earthquake 2005 was rugged mountainous terrain with many
previous mass movements. Most slopes in the epicentral region and along the hanging wall block
of Muzaffarabad Fault from Balakot to Bagh have exposures of weathered bedrock or thin
colluvium. The weathering of rock depends upon the climatic conditions, geology, structure,
topography, vegetation, and slopes of the area. Rain fall is the main weathering agent. Heavy rain
fall, high relief, and highly sheared and fractured rock units made mass wasting an important
degradation process. The degraded talus accumulated on the low foot hills in the earthquake
affected area.
5.1.1. Damages caused by mass movements
The perception of the impact of the mass movements triggered by the Kashmir earthquake 2005
was much lower in comparison with major earthquakes in the world (Harp et al., 1981; Keefer,
1984; Khazai and Sitar, 2003; Wang and Sassa, 2006 and Qi et al., 2010). However, the economic
losses and casualties were much greater than commonly recognized. The Kashmir earthquake
induced mass movements caused damages and devastation throughout the affected area. The
economic value of damages of infrastructure in PAK exceeds comparatively the other part of the
affected region. The mass movements caused sewere damage to cultivated land and destroyed
many structures. Consequently, Muzaffarabad and many other towns and villages leaving isolated
as long as weeks to months, due to the disruption of the communication links. The numbers of
fatalities associated with these mass movements in this earthquake was approximately 26,000
indirectly (Petley et al., 2006) and 1,000 directly (Kamp et al., 2008). The only one big event, the
Hattian Bala rock avalanche killed 575 people and damages five villages.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
47
5.1.2. Earthquake induced mass movements
The Muzaffarabad district, covering an area of about 1,299 km2, was selected as study area
because it was the most affected by the mass movements during the Kashmir earthquake 2005
(Fig. 5.1). Generally, the mass movements were focused in a narrow band clustering in few
specific zones along the reactivated Muzaffarabad Fault (Owen et al., 2008), but it quickly
dissipated with distance away from the fault rupture zone. Similar patterns were recorded in
previous earthquakes triggering mass movements in the world (Bull et al., 1994; Keefer, 2000;
Gallousi and Koukouvelas, 2007).
In general, the mass movements started at an elevation between 800-1,500 meters. The slope angle
lies mainly between 20-60 degrees. The average width of the mass movements ranges between 50-
1500 meters. The outstanding features of these mass movements were large scarps and shallow
slope failures along the main roads, rivers and local streams. However, large scars did not produce
large volumes of mobilized material during the rainy season after the earthquake. In many cases,
rock mass converted into yield debris with small volume. In other hand, some cases the materials
have remained on the hillside with extensive tension cracks and fissures. In addition, many cracks
produced by the earthquake fully developed mass movements so the total mass movement damage
would be much greater as preliminary estimated.
Most of the mass movements in the study area occurred in weathered shale, siltstone, interbedded
sandstone and claystone of the Miocene Murree Formation, and highly fractured carbonates rock
of the Cambrian Muzaffarabad Formation. However, the other formations such as the Precambrian
Hazara and Tanol formations, the Cambrian Mansehra Granite, Paleocene-Eocene sequence and
Quaternary sediments also produced the mass movements in the affected area. Previous studies in
the other part of the world had identified these rocks as highly susceptible to earthquake induced
mass movements, including weakly cemented rocks, artificial fills, uncemented alluvial materials,
and pre-existing mass movements deposits (Keefer, 1984; Khazai and Sitar, 2003; Chigira et al.,
2010). Similar is the case for the Kashmir earthquake triggered mass movements in Muzaffarabad
city, Jhelum valley and Neelum valley areas. In many cases the geological formations that
produced abundant mass movements in this earthquake had been previously identified as highly
susceptible to the mass movements (Farooq, 1997).
The large majority of the mass movements generated throughout the affected area within all types
of geological units were shallow debris cover. All of these geological units are exposed along the
reactivated Muzaffarabad Fault and close to the epicentral area where rock masses are extremely
fractured, and ground motion was very high during the earthquake.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
48
Fig. 5.1 Map showing the distribution of mass movements triggered by the Kashmir earthquake
2005, in Muzaffarabad and surrounding areas. The investigated area is marked by a rectangular
polygon (map of Muzaffarabad district digitized and modified after the map from Planning and
Development department AJK, 2007).
5.2. Terminology and classification of mass movements
Several terminologies and definitions of the mass movement classification system of Varnes
(1978, 1984), Hutchinson (1968, 1988), Keefer (1984), Cruden and Varnes (1996) and Hungr et
al., (2001) are widely adopted in the existing scientific literature. These classifications are
inconsistent with each other, depended on the choice of the scientist who investigated the mass
movements, or who supported their research work. Consequently, different terminologies and
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
49
definitions of mass movement types were observed. Furthermore, the use of these terms differs
substantially in different countries and languages. The most widely accepted classification systems
are the mass movement classification of Varnes (1978), Cruden and Varnes (1996), Keefer (1984)
and Hungr et al., (2001), which can be universally understood. These classification systems are
based on the morphology, mass movement mechanism, geometry of failure area, movement type
and rate, type of material, volume deposit and activity.
5.2.1. Classification system of Varnes (1978)
The classification system of Varnes (1978) is used by many scientists for the study of the types of
failure and the identification of mass movements. In general, various types of mass movements
can be differentiated based on type of movement and type of material (Varnes, 1978). A
classification system based on these parameters is shown in Table 5.1.
Table 5.1 Mass movement classification (Varnes, 1978).
TYPE OF MOVEMENT TYPE OF MATERIAL
BEDROCK ENGINEERING SOILS
Predominantly
coarse
Predominantly
fine
FALLS Rock fall Debris fall Earth fall
TOPPLES Rock topple Debris topple Earth topple
SLIDES ROTATIONAL Rock slide Debris slide Earth slide
TRANSLATIONAL
LATERAL SPREADS Rock spread Debris spread Earth spread
FLOWS Rock flow
(deep creep)
Debris flow Earth flow
(soil creep)
Complex Combination of two or more principal types of movement
5.2.2. Classification system of Cruden and Varnes (1996)
The first classification system of mass movements was proposed by Varnes in 1958, subsequent
modifications have been carried out in 1978 and 1996, which are accepted now universally.
Cruden and Varnes (1996) define that the criterion for the identification of mass movement is the
type of movement, whereas further subdivision is made based on the type of material. The type of
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
50
material is divided into five classes, whereas material is divided into two types, the latter is further
subdivided into debris and earth (Table 5.2).
Table 5.2 Mass movement classification (Cruden and Varnes, 1996).
TYPE OF
MOVEMENT
BEDROCK
TYPE OF MATERIAL
ENGINEERING SOILS
PREDOMINANTLYCOARSE PREDOMINANTLY SOIL
Fall Rock fall Debris fall Earth fall
Topple Rock topple Debris topple Earth topple
Slide Rock slide Debris slide Earth slide
Spread Rock spread Debris spread Earth spread
Flow Rock flow Debris flow Earth flow
5.2.3. Classification system used for this study
In order to classify and define mass movements in this study, the terminology and definitions
adopted in the existing literature were used. The general terminology and classification used is
based on Varnes (1978). The classification used for the statistical analysis of the empirical models
is based on a volumetric nomenclature (Varnes, 1978 and Fell, 1994) to describe the mass
movement travel distance and Fahrböschung angle.
5.3. Types of mass movements induced by the earthquake
The mass movements triggered by the Kashmir earthquake ranged in size from little rock falls upto
a rock avalanche of about 98 million cubic meters. However, the largest numbers of mass
movements were shallow failures of the uppermost few centimeters to meters material. Often only
the debris cover of the hard rocks had been mobilized. Only a few mass movements were deep
seated. The mass movements are mainly divided into two major categories.
i) Shallow mass movements on very steep slopes
ii) Deep seated mass movements
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
51
5.3.1. Shallow mass movements on very steep slopes
The large majority of the mass movements are shallow mass movements with a thickness of less
than 3 meters. They consist of dry, highly disaggregated, and fractured material that fell down the
slope to flatter areas at or near the base of steep slopes (Fig. 5.2).
Fig. 5.2 Multiple shallow failures on very steep slopes in the Hanging Wall Block (HWB)
Muzaffarabad Fault (MF), northeast of Muzaffarabad city. The Muzaffarabad Formation lies in
the HWB and the Murree Formation in the Foot Wall Block (FWB).
Even though relatively small in thickness, the shallow mass movements contributed significantly
to the earthquake related damage. Furthermore, they posed great threat to main roads and
structures at the slope bases. The shallow mass movements are strongly related on strong ground
motion during the earthquake. However, they appear in all geological units because the main
mobilized material was the debris cover. In Muzaffarabad city, the shallow mass movements
stripped off the mountain vegetation cover from the steep slopes. Consequently, many areas totally
denuded of vegetation could be seen in SPOT satellite image (Fig. 5.3). These types of mass
movements were also widely spread in Jhelum and Neelum valley areas along the roads and rivers.
Furthermore, the unstable nature of the debris and presence of the disrupted rock masses along the
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
52
slope above the roadway during the rainy season made reconstruction efforts difficult and the
roads remain closed for long times. In general, mass movements in talus deposits in mountain
areas and soil collapses along the streams in the populated areas were also observed. They caused
less direct damage for settlements. The soil collapses to mainly lie on steep cliffs and have a height
of approximately 50 to 90 meters and slope angles of 70 to 85 degree (Kausar, 2008).
Fig. 5.3 SPOT-5 (2.5 m) image showing the mass movements induced by the Kashmir earthquake
2005 denudated the vegetation on the steep slopes in the north and northwestern part of
Muzaffarabad city, and around the Muzaffarabad Hills.
5.3.2. Deep seated mass movements
The deep seated mass movements induced by the earthquake were far less numerous than shallow
mass movements. In contrast to the shallow mass movements that occurred at the steep slopes, the
deep seated mass movements were observed to be reactivated old mass movements. They were
located on the hanging wall block of the reactivated Muzaffarabad Fault and close to the epicenter
region. The biggest deep seated mass movement associated by this earthquake was the Hattian
Bala rock avalanche, which occurred in the southeast of the study area (see detail in section 5.6.2).
The other three significant large scale mass movements are: Langarpura and Neelidandi rock falls,
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
53
and Panjgran slump and rock fall located in Muzaffarabad city, Jhelum valley and Neelum valley
areas (see details in section 5.6.3, 5.6.4 and 5.6.5).
5.4. Mass movement identification and classification
The total number of mass movements identified by remote sensing technique was more than 2,400
(Sato et al., 2007) for the whole area affected by the earthquake. Among of them, 1460 mass
movements were identified within the study area using SPOT satellite images (Fig. 5.1). Following
that, 127 mass movements were investigated and documented directly in the field by the author.
Five types of mass movement such as landslide, rock fall, debris fall, rock avalanche, and rock fall
and slump have been classified based on the type of movement and type of material as proposed
by Varnes (1978) (Fig. 5.4; Table 5.3). In general, most of the mass movements were smaller in
volume and scale, and a few were large and deep seated. Furthermore, the general characteristics
of the mass movement were observed to be common to those observed in many previous large
earthquakes in other parts of the world (Keefer, 1984; Khazai and Sitar, 2003; Chigira et al.,
2010).
Table 5.3 Types of mass movements examined in the field (Classification after Varnes, 1978).
Types of mass movements Rock Falls Debris Falls Landslides Rock
Avalanches
Slump and
Rock Fall
Number of mass movements 64 55 6 1 1
The term mass movement (Brunsden, 1984) is used in this study to describe all types of mass
movements. The mass movements investigated during field can be classified within five major
types: (a) Landslide (b) Rock fall (c) Debris fall (d) Rock avalanche (e) Slump and rock fall. A
brief account of these types of mass movements is given below.
5.4.1. Landslides
The term landslide can be defined as “down slope movements of soil or rock masses along the well
defined surface of rupture called slip or shear surface (Cruden and Varnes, 1996). They were
identified by the movement of a relatively intact slide mass above a failure surface (Cruden and
Varnes, 1996). The failure surface usually develops at the contact between the border of loose rock
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
54
and bedrock. Generally, landslides are divided into two major types: rotational landslides and
translational landslides.
Fig. 5.4 Map showing the distribution and the types of mass movements investigated during field
surveys in 2009 and 2010 for this study.
5.4.1.1. Rotational landslides
In rotational slides, the sliding movement occurs along a shear surface which is concave upwards
in the direction of the movement whereas, the displaced mass rotates about an axis which is
parallel to the slope (Cruden and Varnes, 1996). The displaced material moved downward beyond
the rupture of the surface to deposit the material at the toe.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
55
5.4.1.2. Translational landslides
Translational landslides or planer landslides displaced the material down slope on a largely planar
surface. In rock slides usually they occur along discontinuities such as bedding planes and joints,
whereas in debris slides the failure can occur along the shallow shear surface (Cruden and Varnes,
1996). If the overlaying material moves as single or less deformed mass, it is called blockslide.
5.4.1.3. Occurrence of landslides in the study area
Several types of mass movements were triggered by the Kashmir earthquake 2005 such as
rotational landslide, debris slide, rock slide and slump. The major types of mass movements in the
study area are translational and rotational landslides. Translational landslides are seen in many
places and are caused by fragile, highly fractured and jointed rock (Fig. 5.5). The movements are
controlled by large weak zones within the structure of the slope-forming material. These
discontinuities include intersecting joint surfaces, inclined or sub horizontal bedding planes, faults,
thrusts and deposits which display variations in shear strength, as well as shear surfaces generated
through the soil by the failure. Rotational landslides are frequently observed in the study area in
lithologies where relatively homogenous material or jointed rocks were present, for example in
Tertiary rocks.
In total 6 landslides were identified and classified during the field survey (Fig. 5.4; Table 5.3).
Most of the landslides have a distinct toe at the base of the hill side. The material involved is
mostly colluvium and weathered portions of densely fractured rock masses.
Fig. 5.5 Transitional landslide in
Botha area, in the northeast of
Muzaffarbad city triggered by the
earthquake.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
56
5.4.2. Rock falls
The rock fall starts with the detachment of rock or cliff from the steep slope along the surface, on
which little or no shear displacements take place. The separation occurs along discontinuities, such
as the fractures, joints and bedding planes. As a result, the material moves down by free fall,
bouncing and rolling (Varnes, 1978). The individual rock falls were identified based on the abrupt
movement of rock masses or rocks and boulders that become detached from steep slopes or cliffs
(Cruden and Varnes, 1996).
5.4.2.1. Occurrence of rock falls in the study area
The predominant type of observed mass movements in the study area is rock fall (Fig. 5.6). Rock
falls are about 71 % of the total mass movements triggered by the earthquake in the whole affected
area (Owen et al., 2008). As a result they blocked the main roads and the connecting roads from
Muzaffarabad city to other areas, and damaged the public utilities during the earthquake. Most of
the rock falls occurred along the Neelum valley and Jhelum valley roads and caused numerous
casualties during the incident.
In total 64 rock falls events were identified and classified according to the classification system of
Varnes (1978) during the field investigation (Fig. 5.4; Table 5.3). They were identified at many
locations along the road and steep cliffs in Muzaffarabad city, Jhelum valley and Neelum valley
areas. In the area of Neelum valley and Jhelum valley, the rock falls mostly occurred in jointed
Fig. 5.6 The rock fall occurred in weathered shales and
fractured sandstones of Miocene Murree Formation
near Batmang area along the main road of Neelum
valley. This rock fall blocked the Neelum valley road
many days after the earthquake.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
57
sandstones of Miocene Murree Formation on steep slopes having slope angles of more than 50
degrees. However, in Muzaffarabad city and the surrounding area the rock falls occurred in
fractured carbonate rocks of the Cambrian Muzaffarabad Formation along the hanging wall block
of the reactivated Muzaffarabad Fault. In most of these locations, the rocks are highly fragile,
fractured, detached and some rock blocks are still hanging. The most common morphological
indicator of the rock falls in the study area is the accumulation of rock falls material and talus
deposit at the foot of the slope (Fig. 5.6).
5.4.3. Debris falls
The debris falls are similar to rock falls, except they involve a mixture of soil, regolith, vegetation,
and rocks. The movement is relatively free downward or forward falling of an unconsolidated or
weathered portion of densely rock masses from cliff or steep slopes.
5.4.3.1. Occurrence of debris falls in the study area
The earthquake triggered widespread debris falls occurred on steep slopes along the roads, rivers,
stream cuts and low altitude areas (Fig. 5.7). This phenomenon is more frequently observed along
the link roads, due to opening of the fracture as a result of extensive and uncontrolled blasting
during the construction of these roads.
The total 55 events of debris falls were identified from different locations in Muzaffarabad city,
Jhelum valley, and Neelum valley areas (Fig. 5.4; Table 5.3). In most of the locations they become
detached from steep surfaces, where loose or unconsolidated material was present. These locations
are mainly associated with slope angles of 20-60 degrees. They are recognized by common
morphological indicators such as typical slope morphology and grain size distribution.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
58
5.4.4. Rock avalanches
The term rock avalanche, also called Sturzstrom (Heim, 1932) proposed by Hungr et al., (2001)
and following closely that of Varnes (1958), is used as the general term for a failure of this type. It
involves unconfined shallow flow of fragmented rock on a steep slope. Rock avalanche originates
as a mass of rock in a rockslide or rock fall fragmenting during failure, finally flowing large
masses of soil, debris, rock or a mixture of these materials in response to the force of gravity. They
are triggered by such events as an earthquake tremors or excessive rainfall on high gradient slopes,
often where materials are loosely consolidated, weathered, or highly fractured.
5.4.4.1. Occurrence of rock avalanche in the study area
The largest mass movements associated with this earthquake was the Hattian Bala rock avalanche,
with an average thickness approximately 60 meters, was truly catastrophic and was responsible
most of the loss of life, caused by the mass movements (see detail description in section 5.6.2).
5.4.5. Slump and rock fall
Slump consist a mass of soil or rock material sliding along a curved surface. The material involved
in the slump rotates along the failure surface. It occurs usually in an area where rock units are
unconsolidated or weak rock layers. Sometime there is combination of slump and rock fall.
Fig. 5.7 Debris fall located in Dunga Kas
Nala in Neelum valley area triggered by the
Kashmir earthquake 2005. This debris fall
was occurred at low altitude area along the
stream.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
59
5.4.5.1. Occurrence of slump and rock fall in the study area
The only one slump and rock fall was identified in the study area. It occurs in Neelum valley area
close to the earthquake epicenter (Fig. 5.4; Table 5.3). This was an old mass movement which
reactivated during the earthquake (see detail description in section 5.6.5).
5.5. Geographic distribution of mass movements in the study area
For the geographical distribution of mass movements, the study area is outlined into three main
parts based on the geographical and geological significance (Fig. 5.8).
.
Fig. 5.8 Map showing the location of the 3 main study areas: 1. Muzaffarabad city, 2. Jhelum
valley, 3. Neelum valley.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
60
1. The Muzaffarabad city area, which encompasses approximately 82 km2 between Chhater on the
south and Kamsar on the north (Fig. 5.10).
2. The Jhelum valley area which covers approximately 392 km2 between Subri on the southwest
and Hattian Bala on the east along the Jhelum river (Fig. 5.15).
3. The Neelum valley area covers approximately 315 km2 from Gujju Saidan on the northwest to
Nausada on northeast along the Neelum river (Fig. 5.17).
These three study areas cover a total of about 790 km2, which is about 65 % of the area. The mass
movement distribution maps were prepared, using SPOT satellite imageries and ground based field
work conducted in 2009-10 for this study. In addition, mass movement inventories containing
information of geometrical parameters were prepared. These inventories provide the information
about length, width, area, and volume of each mass movement.
5.5.1. Mass movements in Muzaffarabad and surrounding area
Muzaffarabad city is located at the confluence of Jhelum and Neelum rivers (Fig. 5.8). The old
part of the city is situated on the left bank of the Neelum river on a Quaternary terrace of river bed
deposits. The hills around the city have mostly gentle to moderate slopes in southwest. In the north
of the city the slopes are mostly steep. Sedimentary rocks of Precambrian to Tertiary age are
exposed in the area. Tectonically the area is very sensitive because two major faults (Muzaffarabad
Fault and Jhelum Fault) run through the area and make a knot just in the NW of the city (Fig. 5.8).
Due to intensive tectonic activities all the strata is highly sheared and fractured. Many small folds
and faults have been observed throughout the area.
Muzaffarabad city was the most affected by the Kashmir earthquake 2005. Some parts of the city
were badly damaged and numerous mass movements occurred on the steep slope of Muzaffarabad
hills, around Muzaffarabad city (Fig. 5.9). These mass movements affected a large number of
communities and most of the structures were totally destroyed or damaged. The inhabitants of the
Muzaffarabad city pose the major threats due to these mass movements.
The 250 mass movements covering an area of 82 km2 of Muzaffarabad city were identified by
SPOT images, taken immediate after the Kashmir earthquake 2005 (Fig. 5.10). Based on these
SPOT satellite images, 60 mass movements, associated with the reactivated Muzaffarabad Fault
and Jhelum Fault were investigated by detail field mapping on the scale of 1:10,000 during the
field trips in 2009 and 2010 (Fig. 5.11).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
61
Fig. 5.9 Quickbird (0.6 m) image showing shallow and deep mass movements on steep slopes of
Muzaffarabad hills, around Muzaffarabad city. Outline shows the boundary of the mass
movements.
All of the mapped mass movements have the following common features: a source area which
comprises the main scarp, a mass movement trajectory, and a toe or deposition fan, where the mass
movement is accumulated. It should be noted that the deposited fan might not be well developed
because the mass movement materials were deposited on the channel, or it has been eroded by
rivers and local streams during the seasonal water level rises. In addition, many mass movements
were complex to distinguish and to demarcate the boundaries from the other ones (Fig. 5.9).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
62
Fig. 5.10 Mass movement distribution map of the Muzaffarabad city and surrounding area.
The inventory map of 60 mass movements of Muzaffarabad city and surrounding area was
prepared entirely from the ground based field work (Table 5.4). The mass movements were
classified in the field using the criteria proposed by Varnes (1978). The mass movements include
rock falls, debris falls and landslides (Fig. 5.11), which are concentrated on the steep slope,
carbonate rocks of Muzaffarbad Formation, weathered shales, claystones and siltstones of Murree
Formation, and slates of Hazara Formation.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
63
Fig. 5.11 Types of mass movements in Muzaffarabad city and the surrounding area.
The mass movements were initiated at an elevation of 820-1350 m. The relative elevation
difference from Neelum river to the top of the ridge is about 800 m. The smallest mass movement
mapped in the area was 0.004 km2 and located at Lohargali area, while, the largest mass movement
area was 0.610 km2 (Table 5.4) located at Neelidandi along Muzaffarabad Fault (see detail in
section 5.6.4). The distribution of length, width, area and volume is shown in Table 5.4. For the
examined mass movements the length varies between 40 m to 650 m, with a mean value of 305 m,
and maximum width ranges between 25 m to 1,370 m, with a mean value of 200 m (Table 5.4).
The smallest volume of the mapped mass movements was 0.002 million m3 while, the largest mass
movement has a volume of 3.1 million m3, estimated for Neelidandi rock fall (Table 5.4).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
64
Table 5.4 Type, length, width, area and volume of mass movements in Muzaffarabad city and the
surrounding area.
ID Mass movement type
Length (m) Width (m) Area (km2) Volume (million m3)
1 Rock Fall 486 1,370 0.610 3.1
2 Rock Fall 416 455 0.162 0.108
3 Rock Fall 650 270 0.139 0.250
4 Rock Fall 110 800 0.127 0.044
5 Rock Fall 330 380 0.098 0.25
6 Rock Fall 625 200 0.092 0.125
7 Debris Fall 600 300 0.089 0.36
8 Debris Fall 300 360 0.087 0.054
9 Rock Fall 230 170 0.079 0.088
10 Debris Fall 248 290 0.077 0.06
11 Rock Fall 590 120 0.076 0.141
12 Rock Fall 306 265 0.071 0.42
13 Rock Fall 520 110 0.065 0.060
14 Debris Fall 160 260 0.060 0.041
15 Rock Fall 360 220 0.059 0.075
16 Rock Fall 392 160 0.058 0.156
17 Rock Fall 254 140 0.056 0.06
18 Rock Fall 288 210 0.053 0.102
19 Rock Fall 350 135 0.051 0.105
20 Rock Fall 436 160 0.049 0.12
21 Rock Fall 258 230 0.047 0.04
22 Debris Fall 364 140 0.044 0.069
23 Rock Fall 206 240 0.044 0.112
24 Debris Fall 300 165 0.043 0.024
25 Landslide 420 115 0.041 0.097
26 Debris Fall 270 170 0.041 0.036
27 Rock Fall 405 100 0.039 0.018
28 Rock Fall 446 80 0.037 0.12
29 Rock Fall 258 155 0.036 0.138
30 Rock Fall 366 100 0.034 0.044
31 Debris Fall 415 130 0.030 0.035
32 Rock Fall 330 85 0.029 0.072
33 Debris Fall 174 250 0.029 0.022
34 Rock Fall 168 100 0.023 0.066
35 Rock Fall 270 100 0.023 0.018
36 Debris Fall 190 130 0.023 0.036
37 Rock Fall 218 100 0.021 0.008
38 Debris Fall 332 90 0.018 0.018
39 Rock Fall 185 120 0.018 0.007
40 Debris Fall 200 130 0.018 0.013
41 Debris Fall 386 75 0.017 0.012
42 Rock Fall 282 105 0.017 0.03
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
65
43 Rock Fall 470 30 0.016 0.007
44 Rock Fall 218 65 0.016 0.009
45 Rock Fall 160 110 0.016 0.012
46 Debris Fall 195 85 0.014 0.016
47 Debris Fall 200 85 0.013 0.009
48 Debris Fall 210 90 0.013 0.018
49 Debris Fall 294 40 0.011 0.014
50 Rock Fall 144 70 0.011 0.015
51 Debris Fall 200 70 0.011 0.007
52 Rock Fall 300 570 0.011 0.066
53 Debris Fall 350 30 0.010 0.008
54 Debris Fall 178 40 0.006 0.002
55 Debris Fall 230 25 0.005 0.002
56 Landslide 40 180 0.005 0.036
57 Rock Fall 110 55 0.004 0.002
58 Landslide 400 180 0.044 0.32
59 Debris Fall 330 780 0.069 0.01
60 Landslide 370 260 0.090 0.13
The clustering of the mass movements has occurred on specific geological units and in specific
geomorphological settings:
1. The primary region affected by the mass movements is the northern and northwestern part of
Muzaffarabad city, the mass movements occurred near the Muzaffarabad Fault, on carbonate
rocks, principally dolomites, dolomitic limestones, and limestones.
2. In the central part, the mass movements are found along the hanging wall block of
Muzaffarabad Fault on the steep slope of dolomitic limestones.
3. In southwestern part of Muzaffarbad city, the mass movements were found at Saman Bandi and
Lohargali areas along the Jhelum Fault and the Jhelum river respectively. In contrast, fewer mass
movements were triggered in the southern part of Muzaffarabad city.
In the northern and northwestern part of Muzaffarabad city (Fig. 5.12), the failures generally
occurred on the ridge exposed in the north and northwest direction, on very steep slopes, highly
sheared and fractured dolomitic-limestones. In addition, the mass movements denuded almost
entire slopes of the ridge, and the concentration of mass movements was very high as compared to
the other parts of the city (Figs. 5.3 and 5.9).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
66
Fig. 5.12 Mass movement failures on the steeper slopes of carbonate rocks in the northern and
northwestern part of Muzaffarabad city.
Virtually, all the mass movements were shallow failures that involved the down slope movement
of talus material and of highly sheared and fractured hard rocks, caused by the strong ground
motion during the earthquake. The predominant modes of failures are rock falls and debris falls.
The geological formation associated with the mass movements in this area is the Cambrian
Muzaffarabad Formation. The Muzaffarabad Formation is composed of cherty and stromatolitic
dolomites with cherty white and grey bands, limestones and black shales. The mass movements
were concentrated near the Muzaffarabad Fault (Fig. 5.10). The dolomites of the Muzaffarabad
Formation along the Muzaffarabad Fault are deformed. All observed shallow mass movements
were generated during the earthquake.
In the central part of Muzaffarabad city the vast majority of mass movements are visible in the
dolomitic limestones of the Muzaffarabad Hills (Fig. 5.13). The mountains behind the
Muzaffarabad city are formed from Cambrian limestones and dolomites, whilst the more gentle
slopes in the front are formed mostly from colluvium (deposits of old mass movements and other
slope forming processes). The material exposed in the area is composed of predominantly talus,
big blocks of limestone breccia and in the lower parts of shales and highly fractured and sheared
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
67
sandstones of the Murree Formation. The average thickness of talus is less than 1 m. The
Muzaffarbad Fault is passing through the middle of the area (Figs. 5.10 and 5.11). The length of
the affected slope is almost 600 m and the width is 1400 m. The general angle of slope is up to 45
degrees. The slopes are dipping towards NW.
The bottom of the mass movements is marked by Muzaffarabad Fault, where the slope angle
decreases and the slopes are vegetated and inhabited. The mass movements were concentrated
along the shear zone of Muzaffarabad Fault. This brittle shear zone is 0.5 to 0.75 km wide (Baig,
2006).
Fig. 5.13 Shallow slope failures on the steep slope of Muzaffarabad Hills in dolomitic limestone.
Most of the activated material was the former talus cover.
In the southwestern part of Muzaffarbad city the mass movements were found (Fig. 5.10) as the
cluster at Lohargali area. In addition, a small scale landslide occurred along the Shahwi Katha,
along the Jhelum Fault (Fig. 5.10). Most of the mass movements in this area already existed before
the Kashmir earthquake 2005 and underwent a reactivation during the earthquake. These mass
movements occurred, in the Precambrian Hazara Formation. The Hazara Formation consists of
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
68
slates, phyllites and shales with minor limestones and graphitic layers. The observed mass
movements include rock falls, debris falls and landslides.
Fig. 5.14 Mass movements in the southwest of Muzaffarabad city reactivated by the Kashmir
earthquake 2005.
5.5.2. Mass movements in Jhelum valley area
In the Jhelum valley area from Subri to Hattian Bala, mass movements were identified and
mapped using SPOT images and on ground based investigation along the hanging wall block of
the reactivated Muzaffarabad Fault and Jhelum river (Fig. 5.15). These mass movements occurred
due to the failures of the scarp faces and steep slopes consisting of weathered shales, claystones,
siltstones and fractured sandstones (Fig. 5.16). In addition, pre earthquake mass movements
located at Langarpura, and Kuroli were reactivated during the earthquake. Towards further east the
biggest mass movement associated with this earthquake, the Hattian Bala rock avalanche occurred
along the Muzaffarabad Fault (see detail in section 5.6.2). In addition, large numbers of shallow
mass movements were also found in this region. The mass movements include shallow rock falls
and debris falls on steep slopes along the hanging wall block of the reactivated Muzaffarabad
Fault, whereas few landslides also occurred at Bandi Karim Haider and Dhallah area. Many mass
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
69
movements were observed along the local streams and river terraces. In general, the mass
movements occurred in south and southwest facing direction and there was a slight trend that the
large mass movements occurred on vertically convex slopes rather than on concave slopes (Sato et
al., 2007).
Fig. 5.15 Mass movement distribution map of the Jhelum valley area.
The geological formations associated with these mass movements are the Miocene Murree
Formation and Quaternary sediments. In the Jhelum valley area the Muzaffarabad Fault runs
within the Murree Formation and Quaternary sediments. The Murree Formation is divided into a
lower and an upper part due to different lithology. It varies from undeformed competent beds to
tightly folded and highly fractured strata. Compressional forces are responsible for the occurrences
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
70
of these joints. Joints are generally open, especially in sandstones with weathering surface. The
joint planes are smooth and slightly covered and filled with soft rock.
Fig. 5.16 Debris fall occurred in the red clay of the Miocene Murree Formation in Jhelum valley
area triggered by the Kashmir earthquake 2005, additionally caused by undercutting of the slope
for road construction.
An inventory of 33 mass movements was prepared during field investigation in 2009-10, covering
an area of 320 km2 in the Jhelum valley area (Table 5.5). The mass movements initiated at an
elevation of about 860-2050 m. The smallest mass movement covering an area of 0.002 km2
located at Garhi Dopatta, while, the largest mass movement covering an area of 2.02 km2 is
located at Hattian Bala (Table 5.5). The distribution of length, width, area and volume is shown in
Table 5.5. For the examined mass movements the length vary between 70 m to 2350 m, with a
mean value of 280 m, and maximum width ranges between 30 m to 1,470 m, with a mean value of
190 m. The smallest volume of the mass movement is 0.002 million m3 while the largest mass
movement has volume of 98.0 million m3, estimated for the Hattian Bala rock avalanche (Table
5.5).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
71
Table 5.5 Type, length, width, area and volume of mass movement distribution in Jhelum valley
area.
ID Mass movement type
Length (m) Width (m) Area (km2) Volume (million m3)
1 Rock Avalanche 2,350 1,470 2.02 98.0
2 Rock Fall 805 900 0.61 5.76
3 Debris Fall 160 340 0.40 0.05
4 Rock Fall 800 170 0.10 0.19
5 Rock Fall 248 370 0.064 0.12
6 Rock Fall 298 180 0.053 0.1
7 Debris Fall 400 110 0.040 0.032
8 Debris Fall 90 400 0.040 0.016
9 Landslide 130 200 0.023 0.023
10 Debris Fall 398 40 0.018 0.022
11 Debris Fall 175 250 0.017 0.05
12 Debris Fall 110 220 0.017 0.002
13 Debris Fall 162 110 0.015 0.05
14 Rock Fall 208 80 0.015 0.01
15 Rock Fall 166 110 0.014 0.005
16 Debris Fall 140 110 0.013 0.012
17 Landslide 95 160 0.013 0.02
18 Debris Fall 110 100 0.011 0.01
19 Rock Fall 232 60 0.011 0.002
20 Debris Fall 140 90 0.010 0.006
21 Debris Fall 180 60 0.009 0.003
22 Debris Fall 126 60 0.007 0.006
23 Debris Fall 86 85 0.008 0.008
24 Rock Fall 130 80 0.008 0.005
25 Rock Fall 130 50 0.007 0.006
26 Rock Fall 100 80 0.007 0.005
27 Rock Fall 76 60 0.006 0.007
28 Debris Fall 112 65 0.005 0.004
29 Debris Fall 100 65 0.005 0.004
30 Debris Fall 160 30 0.005 0.002
31 Debris Fall 75 60 0.004 0.007
32 Debris Fall 88 60 0.004 0.005
33 Debris Fall 70 30 0.002 0.002
5.5.3. Mass movements in Neelum valley area
The Neelum valley is V shaped, and situated in the northeast of Muzaffarabad city along the
Neelum river. It has rugged mountains having an average width of 15 km, with a slope range from
30-65 degrees. The epicenter of Kashmir earthquake 2005 was located at Devalian in the Neelum
valley, 18 km away from Muzaffarabad city. During the earthquake, the main road of Neelum
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
72
valley was blocked more than one month due to the massive movements triggered by the
earthquake.
In the Neelum valley section from Gujju Saidan to Nauseri area, a large number of mass
movements were triggered by the earthquake, specifically along the Neelum river. The mass
movement concentration was very high close to the epicentral region (Fig. 5.17). Towards the
further northeast, many mass movements were found close to the MBT and PT. Due to the
remoteness of the area and the difficulty of traveling on very steep slopes, only the mass
movements along the roads and along the Neelum river were investigated for detailed mapping
(Fig. 5.17). The predominant types of mass movements were identified as rock falls and debris
falls.
Most of the mass movements have evidence of the main scarp, which is where the mass movement
originated, a path or channel moved downslope, but the depositional zone at the base of the steep
slope on the valley bottom has been transported during the seasonal water level.
The rock falls and debris falls were recorded along the high altitude mountain ridges in Neelum
valley area (Fig. 5.18). These steep rocky ridges are dipping towards the slope and are intensively
jointed due to high stress. The rapid and quick rock falls, and debris falls at various localities have
destroyed many houses and large number of fatalities has been reported.
The geological formations associated in these mass movements are the Murree, Panjal and Tanol
formations. The rocks of the Miocene Murree Formation in this region are predominantly steeply
dipping interbedded sandstones, siltstones and shales, which are vulnerable to the mass
movements. The Precambrian Tanol Formation consists of chlorite-quartz micaschists, chlorite-
biotite metaquartzites, garnet micaschists, graphitic schists and local marbles. The sequence is
intruded by early Cambrian granites. The rock sequence is multiply deformed and metamorphosed.
The rocks are jointed and fractured.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
73
Fig. 5.17 Mass movement distribution map of the Neelum valley area.
The Carboniferous-Triassic Panjal Formation includes basic metavolcanics, quartzofeldspathic
schists and graphitic schists. In an area near the MBT and PT, occur very closely to each other,
separated by metacarbonates, metasediments, quartzites and graphitic schists. The sequence along
MBT is highly fractured, jointed and sheared. These lithologies, brittle structures and steep slopes
controlled the mass movements locally along MBT during earthquake ground shaking. In this
region, particularly old mass movements were reactivated during the earthquake.
The inventory of 34 mass movements was prepared (Table 5.6) based on field work, covering an
area of 315 km2. The mass movements initiated at an elevation of about 860-1500 m. The smallest
mapped mass movement has an area of 0.004 km2, while, the largest mass movement is covering
an area of 0.39 km2 and is located in the Panjgran area close to the epicentral region. The
distribution of length, width, thickness, area and volume is shown in Table. 5.6. For the mass
movement examined the length varies between 120 m to 950 m, with a mean value of 270 m, and a
maximum width range between 40 m to 650 m, with a mean value of 132 m (Table 5.6). The
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
74
smallest volume of the mass movement is 0.002 million m3 while the largest mass movement has a
volume of 6.75 million m3, estimated for the Panjgran rock fall and slump (Table 5.6).
Fig. 5.18 A view of the Nauseri rock fall close to the MBT in the Neelum valley area.
Table 5.6 Type, length, width, area and volume of mass movement distribution in Neelum valley
area.
ID Mass movement type
Length (m) Width (m) Area (km2) Volume (million m3)
1 Slump and Rock Fall
950 650 0.390 6.75
2 Rock Fall 325 360 0.088 0.03
3 Rock Fall 266 240 0.083 0.108
4 Rock Fall 446 190 0.080 0.012
5 Rock Fall 570 140 0.075 0.08
6 Rock Fall 320 120 0.044 0.04
7 Debris Fall 382 125 0.044 0.018
8 Rock Fall 230 170 0.040 0.017
9 Debris Fall 406 100 0.036 0.10
10 Debris Fall 298 140 0.028 0.026
11 Debris Fall 196 170 0.028 0.011
12 Debris Fall 252 130 0.027 0.06
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
75
13 Rock Fall 420 90 0.026 0.035
14 Debris Fall 150 120 0.026 0.01
15 Rock Fall 230 130 0.021 0.06
16 Rock Fall 254 110 0.021 0.003
17 Rock Fall 218 80 0.019 0.009
18 Rock Fall 180 120 0.016 0.012
19 Rock Fall 210 140 0.015 0.03
20 Rock Fall 136 120 0.015 0.014
21 Rock Fall 250 70 0.014 0.1
22 Rock Fall 162 120 0.014 0.028
23 Debris Fall 136 110 0.012 0.014
24 Debris Fall 137 100 0.011 0.007
25 Debris Fall 330 40 0.011 0.003
26 Debris Fall 175 80 0.010 0.007
27 Rock Fall 354 90 0.010 0.005
28 Rock Fall 202 50 0.008 0.012
29 Rock Fall 120 60 0.008 0.002
30 Debris Fall 198 50 0.007 0.009
31 Rock Fall 132 70 0.006 0.01
32 Rock Fall 126 80 0.006 0.04
33 Debris Fall 128 40 0.005 0.006
34 Debris Fall 124 90 0.004 0.003
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
76
5.6. Mass movement case studies
This section describes significant mass movements such as the Hattian Bala rock avalanche,
Langarpura and Neelidandi rock falls, and Panjgran slump and rock fall associated with the
Kashmir earthquake 2005, in the northern part of Pakistan. The Hattian Bala rock avalanche and
Langarpura rock fall occurred in the southeast of Muzaffarabad and Neelidandi rock fall occurred
in the north of Muzaffarabad city along the hanging wall block of the reactivated Muzaffarabad
Fault (Fig. 5.19). However, Panjgran slump and rock fall occurred close to the epicentral region in
the northeast of Muzaffarabad. In this thesis, the following case studies were investigated and
documented during the field trip of October 2009 and September 2010. Furthermore, they were
analyzed and discussed in terms of distribution, behaviour and characteristics. Description
preferably includes the information about morphology, geometry, geology, structure and deposits
of individual mass movements. The locations of these mass movements are shown in Fig. 5.19.
Fig. 5.19 Map showing the locations of the case studies described in text. (map of Muzaffarabad
district digitized and modified after the map from Planning and Development department AJK,
2007).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
77
5.6.1. Previous studies
After the devastating earthquake, a large number of scientists investigated the area to identify and
characterize the mass movements and their impact in this region. The Hattian Bala rock avalanche
was investigated by Harp and Crone (2006), Dunning et al., (2007) and Schneider (2008). A
detailed geological and structural characterization of the Hattian Bala rock avalanche was not
discussed earlier. In this work, the relationship between geology and structure of the Hattian Bala
rock avalanche is analysed in the first time. Therefore, it was geologically and structurally
mapped. However, detailed geological, geotechnical, geomorphological and structural
characteristics of Langarpura and Neelidandi rock falls and Panjgran slump and rock fall are the
first documented examples of large scale mass movements after the Kashmir earthquake 2005 in
the NW Himalayan of Pakistan.
5.6.2. Hattian Bala rock avalanche
5.6.2.1. Introduction to the Hattian Bala rock avalanche case study
The Hattian Bala area is located in the southeast of Muzaffarabad, in the Jhelum valley area, near
the bank of the Jhelum river (Figs. 5.19 and 5.20). The Jhelum river flows from southeast to the
northwest parallel to the Muzaffarabad Fault and turns abrupt towards southward in Muzaffarabad
city. The area of Hattian Bala is generally hilly and mountainous. Most of the area is highly eroded
and characterized by deeply cut ravines and undulating hilly terrains.
Hattian Bala rock avalanche is the largest mass movement associated by the Kashmir earthquake
2005, having a volume of about 98 million cubic meters. It lies on the hanging wall block of the
reactivated Muzaffarabad Fault. It occurred approximately 32 km southeast of Muzaffarabad in a
tributary of the Jhelum river on Danna Hill, close to the town of Hattian (Figs. 5.19 and 5.20). The
massive catastrophic event caused the death of 575 people. It destroyed five villages and more
than 3000 people became homeless (official sources). The mass movement moved in southeastern
direction, created two natural dams on the valley bottom and blocked the water ways of the Karli
and Tung tributaries of the Jhelum river and formed two landslides dammed lakes. The reservoir
of Karli lake reached full capacity of 62 million m3 in April 2007 (Kazuo et al., 2009). In February
2010, Karli dam was overflown due to heavy rain. The overflow resulted in an outburst flood to
the downstream in Hattian town and Muzaffarabad. The flood left one person dead, swept away 50
houses and damaged the Jhelum valley road. The lake was decreased to roughly one third of its
original size.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
78
Previous studies of the Hattian Bala rock avalanche showed, that it was an old rockslide which was
reactivated by the 8th October 2005 earthquake and transformed into a rock avalanche. Dunning et
al. (2007) classified it as rock avalanche according to the classification system of Heim (1932) and
Varnes (1978).
Fig. 5.20 SPOT-5 image of the Hattian Bala rock avalanche. Outline shows the boundary of the
mass movement and blockage of the Karli and Tung tributaries of the Jhelum river.
5.6.2.2. Geological setting of the Hattian Bala rock avalanche
The Hattian Bala area is broadly situated in the eastern limb of HKS. The active fault along the
eastern limb of HKS in the study area is the Muzaffarabad Fault which is the important active
tectonic feature in the HKS (Baig, 2006). The HKS is built up of Precambrian to Tertiary rocks
which are imbricated and folded (Wadia, 1931; Baig and Lawrence, 1987; Bossart et al., 1988;
Greco, 1991). The core of the syntaxis consists of red beds of sandstones, mudstones, shales and
claystones of Miocene age belonging to the Murree Formation (Calkins et al., 1975).
The detailed geological features of the Hattian Bala area have been studied from the GSP and are
summarized in a geological map (GSP, 2008; Fig. 5.21). In the southeast of Muzaffarabad, from
Muzaffarabad city to Hattian Bala area, the Muzaffarabad Fault runs entirely within the Murree
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
79
Formation and the Kamlial Formation or the Quaternary sediments. In general, the area is
composed of Miocene to Quaternary strata. The Miocene layers trend SE-NW with several small
anticlines and synclines. The Miocene layers are mainly composed of interbedded sandstones,
siltstones with shales and claystones. The Quaternary layers consist of alluvial talus and terrace
deposits. The rock sequence shown in Fig. 5.21 includes the early Miocene of Murree Formation,
the late Miocene of Kamlial Formation and Quaternary sediments. As seen in the Fig. 5.21, the
Murree Formation is overlain by the Kamlial Formation in the northeast and southeast of the
Hattian Bala area. It mainly consists of hard, fine grained interbedded sandstones, siltstones with
shales and claystones. The Kamlial Formation lies in the northwest and southwest of the area
which is mainly composed of shales, sandstones with minor siltstone and claystones. In addition,
the Quaternary alluvium is present throughout the Hattian Bala area and overlies the bed rock of
the Miocene strata.
.
Fig. 5.21 Geological map of the Hattian Bala and surrounding area (compiled and modified after
GSP, 2008).
The Hattian Bala rock avalanche lies about 2 km northeast of the Muzaffarabad Fault trace (Fig.
5.19). The slope of Hattian Bala rock avalanche is composed of the lower Murree Formation. The
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
80
lower Murree Formation consists of hard, fine grained interbedded sandstones, siltstones with
shales and claystones. Furthermore, the rocks are highly sheared and fractured because the area
lies on the hanging wall block of the Muzaffarabad Fault zone. The hanging wall block of the
Muzaffarabad Fault contains steep slopes in highly fractured and sheared rock that is highly
susceptible to failure during seismic shaking. The strong motion of the earthquake instantaneously
weakened the brittle shear surface of the detachment zone of Hattian Bala rock avalanche and
caused the coseismic gravity collapse. According to Jibson et al., (2006), the failure in brittle rock
is most sensitive to high accelerations of ground motion.
5.6.2.3. Structural setting of the Hattian Bala rock avalanche
The Hattian Bala rock avalanche is a deep seated earthquake induced mass movement favored by
structurally controlled southeast plunging syncline known as Danna syncline (Fig. 5.22). However,
it also followed the bedding parallel slip and pre-existing synclinal morphology (Figs. 5.23a and
5.23b). The Danna syncline is formed by the folding of the Early Miocene lower Murree
Formation, which is mainly composed of interbedded sandstones, siltstones with shales and
claystones. The siltstones and claystones dominate the sandstones. The northeastern limb of Danna
syncline dips southwest whereas the southwestern limb dips northeast. The strike along the
northeastern limb of the Danna syncline varies from N44ºW to N80ºW while the strike along the
southwestern limb varies from N14ºW to N80ºW (Table 5.7). The dip of the northeastern limb
varies from 35ºSW to 75ºSW whereas the dip of the southwestern limb ranges from 30º NE to
68ºNE. The northeastern limb of the syncline is further folded by the small Dandbeh synclinal and
anticlinal structures plunging southeast (Fig. 5.22). Trend and plunge of the Danna syncline is
22º/120º, 6º/131º, 25º/118º, 12º/104º and 20º/074º. The attitude of the axial plane of the syncline is
N55ºW/80ºNE, N52ºW/90ºNE, N48ºW/86ºNE and N72ºW/56ºNE (Table 5.7).
Table 5.7 Structural data of Danna syncline, Hattian Bala rock avalanche.
Northeastern Limb Southwestern Limb Fold Axis Axial Plane
N70ºW/60ºSW N14ºW/32ºNE 22º/120º N55ºW/80ºNE
N45ºW/35ºSW N60ºW/35ºNE 6º/131º N52ºW/90ºNE
N44ºW/52ºSW N72ºW/65ºNE 25º/118º N48ºW/86ºNE
N80ºW/72ºSW N55ºW/30ºNE 12º/104° N72ºW/56ºNE
N80ºW/75ºSW N80ºW/68ºNE 20º/074° N73ºW/90ºNE
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
81
Fig. 5.22 Structural map showing the southeast plunging synclinal structural failure of Hattian
Bala rock avalanche.
The Danna and Dandbeh synclines are an open, southwest vergent and southeast plunging F1
Himalayan fold. The axial plane of the Danna syncline is northeast dipping which is parallel to the
northeast dipping Muzaffarabad Fault. Muzaffarabad Fault and Danna syncline are pre-earthquake
Himalayan structures. These structures were used for failure during the October 8th 2005
earthquake.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
82
Fig. 5.23 Geotechnical cross profiles (5.23a, 5.23b) showing the pre-earthquake and post-
earthquake situation. Note: the rock avalanche perfectly follows the pre-existing structure of the
Danna and Dandbeh synclines. The third profile (5.23c) is showing the maximum deposit
thickness of the Hattian Bala rock avalanche.
5.6.2.4. Description of the Hattian Bala rock avalanche
The preliminary investigation indicates that the Hattian Bala rock avalanche moved southeast from
Danna Hill (34°08′32′′ N; 73°42′44′′ E, altitude 2,038 m asl), in the Hattian Bala area. As seen in
Fig. 5.24, the mass movement moved from an elevation of about 2,038 m asl and traveled the
distance of ~2.5 km. The mass movement blocked the Karli river at an elevation of 1,307 m asl
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
83
and rose up the other side of the valley at an elevation of 1,400 m asl, forming the Karli dam with
an average length and width of 1,700 m and 400 m respectively. It varies in depth from 150 to 200
m (Figs. 5.24, 5.26 and 5.27).
The Hattian Bala rock avalanche is mainly composed of sandstones, siltstones, shales and
claystones of Miocene Murree Formation. Most of the rock avalanche material was deposited at
the toe, where it formed a huge embankment around the confluence of Karli and Tung rivers. A
nearly planar sliding surface was exposed parallel to the intersection of the bedding surfaces.
Detailed geotechnical and structural maps as well as one longitudinal and three cross profiles (Figs.
5.22, 5.23, 5.27 and 5.28) of the Hattian Bala rock avalanche were prepared to describe the
initiation mechanism and the geological and structural characterization of the Hattian Bala rock
avalanche.
Fig. 5.24 View of the Hattian Bala rock avalanche structurally controlled by southeast plunging
Danna syncline. Photo facing northwest.
The crest of the Hattian Bala rock avalanche is highly cracked (Fig. 5.25). The length of these
cracks is 50-60 m and their width can reach up to 2 m. The northeast to southwest extensional
cracks on the crown of the mass movement are associated with the ground shaking and extensional
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
84
forces parallel to the steep main scarp of the mass movement. In the northwestern part of the main
scarp, cracks mostly parallel to the scarp are also present (Fig. 5.27).
Fig. 5.25 Multiple extensional ground cracks oriented northeast-southwest on the crown of the
Hattian Bala rock avalanche. Photo facing northwest.
The source area of Hattian Bala rock avalanche is composed of shales, siltstones, sandstones and
claystones. Whereas, the source of the mass movement, which is initiated at an elevation of 2,038
m asl (Fig. 5.24) and moved in southeastern direction towards the Karli river. The rock face of the
top area is inclined to nearly vertical. Multiple joint surfaces both parallel and perpendicular to the
slope face were exposed at the mass movement source area.
The geometry of the Hattian Bala rock avalanche shows that the old rockslide (Fig. 5.26) was
favored to form a large scale rock avalanche during the earthquake. The Hattian Bala rock
avalanche is up to 1,470 m wide, 2,350 m long, and 60 m deep in an average (Table 5.8).
The height distance from top to toe is approximately 700 m (Table 5.8). Due to the remoteness and
lack of resources no borehole data was available. In order to estimate the average depth of the rock
avalanche deposit, we used pre-earthquake contour lines derived from SRTM based DEM and
constructed longitudinal and cross profiles in the field. The position of the shear surface was
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
85
estimated with these profiles and the knowledge of the structure of the syncline. The average
thickness of the mass movement deposit was estimated to be approximately 60 meters. Total
volume of the mass movement material was estimated to be about 98 million cubic meters by
multiplying the deposit area with the average thickness (Table 5.8).
Fig. 5.26 Map of the Hattian Bala rock avalanche from 2005. Note: situation of old rock slide and
its position is derived from Schneider (2008).
The mobility of the mass movement is estimated according to the H/L parameters (Fig. 5.28),
which represent the relationship between the height of fall and maximum travel distance, also
called apparent coefficient of friction or Fahrböschung angle (Heim 1932). The Fahrböschung
angle of the rock avalanche is measured as 17º and maximum travel distance is 2,350 m (Table
5.8). The shear surface followed the bedding parallel slip along siltstone, claystone and sandstone
surfaces (Figs. 5.27 and 5.28). The mass movement trajectory from scarp to toe is 2350 m long
(profile A-A′ in Fig. 5.28). The rock avalanche material traveled 400-500 meters beyond the Karli
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
86
valley and also buried many structures along the Karli river. The geological longitudinal and cross
profiles (Figs. 5.23 and 5.28) show the relation between rock type, position of the initial rock slide,
slide plane and the thickness of the different of the rock avalanche.
Table 5.8 Geometric characteristics of the Hattian Bala rock avalanche triggered by the Kashmir
earthquake 2005 in northern Pakistan.
Crown
Elevation
(m)
Toe
Elevation
(m)
Height
(m)
Maximum
length
(m)
Maximum
width
(m)
Fahrböschun
g angle
Surfac
e area
(km2)
Deposi
t area
(km2)
Average
thickness
(m)
Volume
(*106m³)
2038 1,307 700
2,350 1,470 17º 2.02 1.64 60 98.4
Fig. 5.27 Geotechnical map of the Hattian Bala rock avalanche and the locations of geological
longitudinal and geotechnical cross profiles. Frequent GPS measurements were performed during
field work to mapped the geotechnical details.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
87
The Hattian Bala rock avalanche deposit is composed of debris from sandstone, siltstone,
claystone, shale and mudstone of the lower Murree Formation. The total deposit area was
calculated to be about 1.64 km2 (Table 5.8). The material exposed on the surface of the mass
movement deposit comprises mainly of angular rock fragments having fairly uniform distribution
of particle size ranging from large boulders (1-10 m3) to silty and clayey particles (Fig. 5.23c). The
lower part of the deposits along the artificial spillway consists mainly of blocks of fine grained
sandstones spanning a wide range of sizes, some individual blocks measured more than 5 m2 on a
side (Fig. 5.28).
Fig. 5.28 Longitudinal geotechnical NW-SE profile showing pre-earthquake landscape and the
geotechnical situation after the rock avalanche. Note: the mass movement is parallel to the
southeast orientated synclinal plunge direction and the slip surface follows in many parts the dip
direction of the bedding. The mass movement abuts and accumulates against the right steep wall
of the former Karli valley.
The thickness of the toe area generally ranges between 150-200 m. It is thinner near the Karli river
and thicker in southeast direction along the Tung river. The mass movement deposit in the valley
bottom has an irregular surface. It has ridges and depressions of several meters amplitude that
reflect the flowage of the mass. These ridges and depressions make the surface of the deposit
highly irregular.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
88
Fig. 5.29 The deposit of the Hattian Bala rock avalanche is mainly composed of angular rock
fragments.
5.6.3. Langarpura rock fall
5.6.3.1. Introduction to the Langarpura rock fall case study
Langarpura is a small village, located 12 km southeast of Muzaffarabad, near the bank of the
Jhelum river in the Jhelum valley area (Fig. 5.19). Generally, the relief of the area is gentle at the
terraces and steep behind mountain of Langarpura in the northeast. The terrace materials have been
transported by the Jhelum river and include a huge amount of boulders, pebbles and cobbles. They
are deposited tightly at the right bank of the river. Several mass movements occurred along the
right bank of the Jhelum river on the steep slopes along the hanging wall block of the
Muzaffarabad Fault. These mass movements are small and shallow slope failures, however, the
large scale mass movement such as Langarpura rock fall was identified on the hanging wall block
of the Muzaffarabad Fault during the field trip of October 2009 and September 2010 (Fig. 5.30).
At present the mass movement is classified as rock fall according to classification of Varnes
(1978). The volume of the rock fall is estimated to be about 5.76 million m3. The Langarpura rock
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
89
fall caused neither loss of life nor widespread destruction of property, as it occurred in a remote
area. However, it destroyed a local link road connected from Langarpura to Garhi Dopatta town.
Fig. 5.30 SPOT-5 image of Langarpura rock fall in the Jhelum valley area. The red line shows the
location of Muzaffarabad Fault passing through this area. Outline shows the boundary of the mass
movement.
5.6.3.2. Geological setting of the Langarpura rock fall
The Langarpura area lies on the hanging wall block of the reactivated Muzaffarabad Fault, in the
HKS (Figs. 5.19, 5.30 and 5.31). The area is imbricated and folded. The active Muzaffarabad Fault
is the major fault run in the area. The rocks of the area are intensely deformed due to Himalayan
orogeny. The Muzaffarabad Fault runs entirely within the Murree Formation (which is divided into
lower and upper part based on lithological characteristics) and Quaternary sediments in this area.
The lower Murree Formation in the hanging wall block has been thrusted on the upper Murree
Formation along Muzaffarabad Fault and at places buried the trace of the fault due to the
occurrence of the mass movement (Fig. 5.31). The lower Murree Formation consists of
predominant red shales, siltstones, claystones and subordinate sandstones. However, the upper
Murree Formation includes dominantly coarser sandstones and subordinate shales and claystones.
The terraces present in Langarpura, Bandi Tagian and Kuroli areas are the alluvium deposits of
Jhelum river (Fig. 5.30).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
90
Fig. 5.31 The escarpment failure of the Langarpura rock fall on the hanging wall block (HWB) of
the Muzaffarabad Fault. Note: The significant topographic front formed by the hanging wall block
of the Muzaffarabad Fault. The foot wall block (FWB) has been eroded by river undercutting
before the reactivation of Langarpura rock fall. Photo looking northeast.
5.6.3.3. Description of the Langarpura rock fall
The Langarpura rock fall occurred in the southeast of Muzaffarabad along the hanging wall block
of the Muzaffarabad Fault in the Jhelum valley (Figs. 5.30 and 5.31). The Langarpura rock fall
was reactivated during Kashmir earthquake 2005 in southwest Himalayan thrust direction. The
slope failure is associated with the escarpment failure due to earthquake ground shaking, steep
scarp slope and hanging wall collapse. The lower Murree Formation rocks being soft can easily
slide during earthquake ground shaking. However, monsoon rains and seepages and springs play
role for the sliding of shales, claystones and siltstones of the lower Murree Formation. The detailed
geotechnical map and geological longitudinal profile of Langarpura rock fall were prepared on the
basis of field investigation to understand the process of rock fall initiation (Figs. 5.32 and 5.33).
The Langarpura rock fall has a composite scarp which dips 60-70 degree southwest (Figs. 5.32 and
5.33). Extensional cracks are present along the crest of the scarp (Fig. 5.32). The scarp surfaces are
interconnected by gravitational cracks. The southwest escarpment failure is opposite to the
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
91
northeast dipping hanging wall strata of the lower Murree Formation (Figs. 5.32 and 5.33). The
escarpment failure occurred in the southwest direction. The pre-earthquake scarp has been
reactivated along the steep scarp surfaces. The presence of springs, seepages and weathered
claystone material in the middle and upper parts of the scarp helped to initiate escarpment failure.
The source area of Langarpura rock fall is composed of interbedded sandstones, siltstones with
shales and claystones of lower Murree Formation. The lower Murree Formation shows brittle
deformation in shear zones. The source of the mass movement initiated at an elevation of 1250 m
asl at crown and moved in the southwest direction towards the Jhelum river. The Muzaffarabad
Fault is passing through overburden deposits in the center of the mass movement. The rock fall
overburden deposit buried the northeast dipping brittle shear zone of the fault (Fig. 5.31). Where,
fault has developed numerous cracks and fractures.
The geomorphological and geometrical features indicate that it was an old rock fall scarp which
was reactivated during the earthquake 2005. Total surface area of Langarpura rock fall is
calculated 0.61 km2, whereas, the deposit area is about 0.48 km2 (Table 5.9). The rock fall has
about 800 m length, a maximum width of 900 m and an estimated average depth of about 12 m.
The top of the head scarp is at about 1250 m asl. and stripped at the height of 460 m. The
Fahrböschung angle of the mass movement is measured as 30º. The volume of the mass movement
is estimated about 5.76 million m3 (Table 5.9).
Table 5.9 Geometric characteristics of the Langarpura rock fall triggered by the Kashmir
earthquake 2005.
Location
Name
Type Crown
elevation (m)
Length
(m)
Maximum
Width (m)
Estimated
depth (m)
Height
(m)
Fahrböschung
angle
Total
surface area (m2)
Deposit
area (m2)
Estimated
volume (106 m3)
Langarpura Rock Fall
1,250 800 900 12 460 30º 610,000 480,000 5.76
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
92
Fig. 5.32 The geotechnical map of the Langarpura rock fall and the location of the geological
longitudinal profile shown in Fig. 5.33.
The rock fall body is made up of fragmented rocks, alluvium and alluvial fan material. Parts of the
rock fall material are transported during heavy rain fall to the Jhelum river by local streams (Fig.
5.32). The Jhelum river has eroded the foot wall block before the reactivation of Langarpura rock
fall. The rock fall material is exposed at the toe of the rock fall. The rock is disintegrated into rock
blocks and boulders accumulated near the Jhelum river. In addition, rock blocks in a silty and
clayey material were also present at the toe. The Jhelum river reduces the rock fall material at the
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
93
toe through side erosion during seasonal water level raise during heavy rain fall or seasonal
monsoon.
These deposits include boulders, cobbles and pebbles of Panjal volcanics, siwaliks, quartzites and
sandstones embedded in sand, silt and clay matrix. The alluvial fans in these areas include angular
to sub-angular fragments of sandstones and siltstones of the Murree Formation in clayey and silty
matrix.
Fig. 5.33 Geological longitudinal profile of the Langarpura rock fall. Location of the profile is
shown in Fig. 5.32.
5.6.4. Neelidandi rock fall
5.6.4.1. Introduction to the Neelidandi rock fall case study
The Neelidandi rock fall is located in the north of Muzaffarabad city, capital state of AJK (Fig.
5.19). The Muzaffarabad city was badly damaged due to the earthquake 2005 and number of mass
movements took place in Muzaffarabad city coming down from the steep slopes, particularly in the
north of Muzaffarabad. The topography of the area around these mass movements is complicated
by the presence of the fault and surface uplift of the hanging wall block. It occurred in the
upstream of Neelum river near Muzaffarabad (Figs. 5.34 and 5.35). This catastrophic mass
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
94
movement was reactivated during the earthquake and destroyed many houses and blocked the
Neelum river for 5 hours. In addition, the mass movement developed numerous fractures and
cracks in the eastern and southwestern side. The mass movement was mapped geologically and
structurally on scale of 1:10,000 during the field trip of October 2009 and September 2010 (Fig.
5.37). The geological longitudinal profile was prepared to characterize the post-earthquake
topography and the initial geological situation.
Fig. 5.34 SPOT-5 image of Neelidandi rock fall in north of Muzaffarabad city. The rock fall
blocked the Neelum river for 5 hours immediately after the earthquake. Outline shows the
boundary of the mass movement.
5.6.4.2. Geological setting of the Neelidandi rock fall
Muzaffarabad city lies in the HKS in the NW Himalayan of Pakistan. The Jhelum Fault and
Muzaffarabad Fault lies on the western limb of HKS, and are the major tectonic features of the
area. The Muzaffarabad Fault runs northwest-southeast from Balakot to Bagh area (Fig. 5.19). The
hanging wall block of the fault was deformed into the active hanging wall anticline (Baig et al.,
2008). It is deformed and folded due to compression combined with uplifting. In Muzaffarabad
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
95
city, the active Muzaffarabad Fault thrusted the Cambrian Muzaffarabad Formation over the
Miocene Murree Formation. Whereas, the Jhelum Fault emplaces the Precambrian Hazara
Formation over the Miocene Murree Formation (Baig and Lawrence, 1987). The Muzaffarabad
Formation includes cherty dolomites, white and gray cherty bands, stromatolitic cherty dolomites,
black limestones and shales (Baig and Snee, 1995), whereas, the Murree Formation contains
interbedded sandstones, siltstones with shales and claystones. The Neelidandi rock fall lies in the
hanging wall block of the Muzaffarabad Fault (Figs. 5.37 and 5.38). The ruptures and mass
movements on the hanging wall block are associated with the northwest trending and northeast
dipping Muzaffarabad Fault. The active uplift and strong ground shaking along Muzaffarabad
Fault caused the total collapse of houses and triggered Neelidandi rock fall in the north of
Muzaffarabad (Fig. 5.38).
5.6.4.3. Description of the Neelidandi rock fall
Neelidandi rock fall occurred on an undercut slope of Neelum river. The undercutting had caused
numerous rock falls in former times. The new rock fall has been reactivated during the Kashmir
earthquake 2005. It has blocked the river Neelum and formed the check dam. After some time the
dam was broken and water flew downstream in the river. The detailed geotechnical map and
geological longitudinal profile (Figs. 5.37 and 5.39) have been prepared to understand the process
of mass movement mechanism and initiation
The very large scarp can be observed in Figs. 5.35 and 5.37. The scarp area of Neelidandi rock fall
is concave and very steep (Fig. 5.35). The rock fall occurred on a steep scarp plane dipping 60-70
degree northwest. Scarp failure followed the northwest-southeast extension perpendicular to active
northeast-southwest Himalayan tectonic compression.
The scarp is mainly composed of highly sheared and crushed cherty dolomites and limestones. The
slope failure along the scarp occurred due to earthquake ground shaking prepared by river
undercutting in highly fractured rocks. The multiple joints and fracture surfaces are sub parallel
and orthogonal to the slope face. On the back side of the scarp crest, semicircular and parallel to
scarp tension cracks are developed (Figs. 5.36 and 5.37).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
96
Fig. 5.35 The very steep scarp of Neelidandi rock fall in the hanging wall block of Muzaffarabad
Fault (MF). The scarp is formed in the highly sheared cherty dolomite of Muzaffarabad Formation.
The Muzaffarabad Formation lies in the hanging wall block and Murree Formation in the foot wall
block. Photo looking southeast.
The tension cracks are up to 2 m wide. The crest of scarp has extensional cracks dipping 60-70
degrees towards the northwest dipping main scarp of the rock fall. The tension crack zone along
the crest of the scarp varies in width from 50-100 m. This tension crack zone can fail by future
triggering events like earthquakes or extreme monsoon rains.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
97
Fig. 5.36 Northwest dipping and oblique tension cracks on the crown of the Neelidandi rock fall.
The cracks are formed due to earthquake ground shaking and stress release behind the new steep
scarp. These cracks can cause a collapse of material behind the scarp during monsoon rains and
earthquake shocks. Photo looking northeast.
Slopes above the main scarp in the northeast direction are steep, rising into forest with small
farms, many of them have been destroyed. Agricultural terraces have been constructed around
some of these cleared areas. The edge just above the main scarp is broken by a number of arcuate
fissures and small scarps. However, these do not appear to be as extensive as at other sites but
restricted to defining areas of the present scarp which will fail by retrogression of the unstable cliff
face. There are also a number of fissures along the interfluves area directed to both tributary
valleys to the east of the main scarp, reflecting potential failure of these slopes.
The source rock of the mass movement is sheared cherty dolomite-limestone of Muzaffarabad
Formation. The main scarp reached an elevation of 1100 m asl. The escarpment height was 355 m
(Table 5.10). Almost no material remained in the source area.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
98
Total surface area of the Neelidandi rock fall is calculated about 0.61 km2, whereas, rock fall
deposit has an area of 0.20 km2. The length of the rock fall is 480 m. It has a maximum width of
1,370 m. The average estimated thickness is 15 m (Table 5.10).
The estimated volume of the rock fall is about 3.1 million m3, calculated by multiplying the
deposit area with the average thickness. The relationship between the height of fall and maximum
travel distance, also called apparent coefficient of friction (Heim, 1932). The Fahrböschung angle
of Neelidandi rock fall measured directly in field is 36º, whereas, the rock fall travel the distance
of 480 m.
Fig. 5.37 Geotechnical map of the Neelidandi rock fall and the location of the geological
longitudinal profile shown in Fig. 5.39.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
99
Table 5.10 Geometric characteristics of the Neelidandi rock fall triggered by the Kashmir
earthquake 2005.
Location Type Crown
elevation
(m)
Length
(m)
Maximum
width (m)
Estimated
depth (m)
Height
(m)
Fahrböschung
angle
Total
surface
area (m2)
Deposit
area (m2)
Estimated
volume
(106 m3)
Neelidandi Rock fall
1,100 480 1,370 15 355 36º 610,000 207,000 3.1
Fig. 5.38 The hanging wall block of Muzaffarabad Fault is deformed into active hanging wall
anticline (Baig et al., 2008). Note total destruction of houses (Earthquake intensity XI) is due to
strong earthquake ground shaking along Muzaffarabad Fault. Photo looking northeast.
After the main mass movement, the slope of the ridge was highly fractured and unstable that rock
fall continued to occur. The rock fall body includes the rock material of cherty dolomite-limestone.
The rock unit is highly sheared and crushed forming breccia zone varying in width from 500-1000
m. The geological longitudinal profile (Fig. 5.39) shows the relation between the initiation of rock
fall with respect to lithology and structure. The geological and structural parameters (Fig. 5.37)
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
100
show that the earthquake strong ground motion (Intensity XI; Fig. 5.37) weaken the brittle shear
zone in the hanging wall of Muzaffarabad fault and caused the collapse of the Neelidandi rock fall.
The rock mass is composed of cherty dolomite and limestone rocks of the Muzaffarabad
Formation. The rock fall material forms the active cone at the toe of the rock deposit. The size of
the deposit material in scale is less than 1 m3, although occasionally rocks with more than 1 m3
across can be seen. The 15-20 m thick rock fall mass has been significantly eroded by the Neelum
river during heavy monsoon rains and rising of seasonal river water level. The only small part of
the rock fall body forms the cone at the toe.
Fig. 5.39 Geological longitudinal profile of the Neelidandi rock fall. Location of the profile is
shown in Fig. 5.37.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
101
5.6.5. Panjgran slump and rock fall
5.6.5.1. Introduction to the Panjgran slump and rock fall case study
The Panjgran village lies in the Neelum valley area, in the northeast of Muzaffarabad city, where
the Neelum river flows from northeast to southwest (Fig. 5.19). The epicenter of Kashmir
earthquake 2005 was located about 8 km northwest of this village. The area is characterized by
rugged topography, very high relief, and very steep slopes. The topography is prone to the mass
movement due to its unstable conditions of the rock masses. This instability has caused failure on
many mass movements before and during the earthquake in this region. The topographic elevation
changes from 850 m at the Neelum river to 1450 m at the top ridge (Fig. 5.40), where the Panjgran
slump and rock fall was reactivated during the earthquake 2005. The mass movement blocked the
main Neelum valley road for many days. Nearly, 300-400 m road were totally destroyed due to the
reactivation of the slump material at the base. However, no causality or damages of houses was
reported during the reactivation of this mass movement
5.6.5.2. Geological setting of the Panjgran slump and rock fall
Geologically, the area is situated close to the MBT and PT (Fig. 5.19). The Panjal Formation lies
between the MBT and PT. Along the MBT the Panjal Formation has been thrusted over the
Murree Formation (Khan, 1994). The sequence of rocks along the MBT is highly fractured, jointed
and sheared. The lower contact of the Panjal Formation is faulted and upper contact is the MBT
with the Murree Formation (Khan, 1994). The large area of Neelum valley is covered with the
Murree Formation. The main part of the Murree Formation is exposed along the right and left bank
of the Neelum river. The brittle structure of the rocks and the steep undercut slope controlled the
triggering of the mass movement near MBT during earthquake. However, most of the mass
movements were occurred in this area due to the debris cover. The Panjgran slump and rock fall is
made up of the Miocene Murree Formation consisting of interbedded sandstones, siltstones with
shales and claystones. The shales and claystones are predominant; whereas sheared sandstones are
subordinate. Slope of the mass movement dips in opposite direction to the dip direction of bed
rock.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
102
Fig. 5.40 SPOT-5 image showing the location and boundary of the Panjgran slump and rock fall
occurred in the northeast of Muzaffarabad, in the Neelum valley area. Outline shows the boundary
of the mass movement.
5.6.3.4. Description of the Panjgran slump and rock fall
The Panjgran slump and rock fall is located 36 km away in the northeast of Muzaffarabad city,
close to the epicenter of Kashmir earthquake 2005, in the Neelum valley area (Figs. 5.19 and
5.40). It is a major ancient mass movement which was reactivated during the Kashmir earthquake
2005. The mass movement occurred on Panjkot ridge (34° 25′ 47′′ N; 73° 37′ 12′′ E, altitude 1,450
m asl) and moved in northeast direction towards the Neelum river (Fig. 5.41). Prior to the
triggering of the mass movement during the earthquake, the Neelum river had frequently undercut
and oversteepened the slope. This reduced the overall stability of the slope on hill side.
The slope failure is associated with the escarpment failure (Fig. 5.41). Beside the material and
slope steepness the under construction of main Neelum road at the base of the mass movement was
the major driving force for this escarpment failure. The mass movement was mapped on scale of 1:
10,000 during the field trip for this study. Detailed geotechnical map and geological longitudinal
profile was prepared to understand the mechanism and initiation process of the mass movement
(Figs. 5.42 and 5.43).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
103
The crown of the mass movement above the main scarp is steep, with small agricultural farms.
Residential houses and agriculture terraces have been constructed around the scarp and crest of the
mass movement and thick forest is present in the northeast side of the scarp. The cracks parallel to
the scarp are appeared in the western part of the mass movements (Fig. 5.41). They are 1 to 5 m
long, 8-12 cm wide, and nearly less than 1 m in deep. The scarp of the mass movement is
composed of fractured sandstones, siltstones, shales and claystones of Miocene Murree Formation
(Fig. 5.42). The height of the scarp varies from 30 m in the western part of the scarp to 200 m at
the top of the Panjkot ridge. The shape of the scarp is circular and it dips into the hill side towards
the Neelum river.
Fig. 5.41 An overview of the Panjgran slump and rock fall which occurred close to the epicentral
region of the Kashmir earthquake 2005. Note: the position on an undercut slope of Neelum river.
Fig. 5.42 shows the geometry of the Panjgran slump and rock fall. Total surface area of the mass
movement is calculated 0.39 km2 (Table 5.11). The mass movement initiated at an elevation of
1450 m asl at the top of the ridge and stripped at the height of 600 m. The mass movement has
about 950 m length and a maximum width 650 m. The estimated average depth is 25 m. The
volume of the mass movement is roughly estimated about 6.75 million m3. The Fahrböschung
angle is measured about 35º (Table 5.11).
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
104
Table 5.11 Geometric characteristics of the Panjgran slump and rock fall triggered by the Kashmir
earthquake 2005.
Location
Name
Type Crown
elevation
(m)
Length
(m)
Maximum
Width (m)
Estimated
depth (m)
Height
(m)
Fahrböschung
angle
Total
surface
area (m2)
Deposit
area (m2)
Estimated
volume
(106 m3)
Panjgran Slump and rock fall
1,450 950 650 25 600 35º 390,000 278,000 6.75
Fig. 5.42 Geotechnical map of the Panjgran slump and rock fall and the location of the geological
longitudinal profile shown in Fig. 5.43.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
105
It is a rotational mass movement on northeast facing slope in which strata dip opposite to the hill
side. The initial slope movement involved the slumping in weathered, jointed shale and sandstone
at the foot of the mass movement. On the higher part of the mass movement the scarp face is
exposed where rock fall material detached from the bed rock and moved down slope. At the foot
of the scarp the old slump mass is present which extends the full width of the mass movement. The
slump is characterized by the rotational movement and is mostly covered by the rock debris,
derived from the rock fall that triggered during the earthquake. The lower part of the slump, below
the main road is characterized by steep slope with slope angle of upto 50º (Fig. 5.42).
The Panjgran slump and rock fall deposit has an area of 0.278 km2 including slump and rock fall
material (Table 5.11). The surface of the deposit comprises sandstones, siltstones, shales and
mudstones. The size ranges from sand to large boulder with many very angular blocks >1 m in
diameter. The material at the toe of the mass movement was transported by the Neelum river
during seasonal water level rises.
Fig. 5.43 Geological longitudinal profile of Panjgran slump and rock fall. Location of the profile is
shown in Fig. 5.42.
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
106
5.6.6. Conclusions based on the case histories
The Hattian Bala and other three large scale mass movements show that lithology, structure and
geometry are three important factors contributing to trigger these mass movements during the
Kashmir earthquake 2005. Geometry and failure mode of Hattian Bala rock avalanche was
strongly controlled by tectonics and lithology, bedding parallel slip and southeast plunging
synclinal structures, and the pre-existing rock slide are the main features. The various aspect of the
Hattian Bala rock avalanche were analyzed as follows: The reactivation of Hattian Bala rock slide
on the hanging wall block of the Muzaffarabad fault was the result of the ground shaking,
structural failure, hanging wall collapse and escarpment failure. The Danna and Dandbeh synclines
were formed by the Himalayan F1 folding of the Murree Formation. Claystones, siltstones and
subordinate sandstones of the lower Murree Formation are prone to mass movements due to
inclined layering. The Danna and Dandbeh southeast plunging synclinal structural failure followed
the bedding parallel slip along the bedding planes of claystones, mudstones and sandstones. In
general, earthquake deformation contributed in a co-seismic gravity collapse of the Hattian Bala
mass movement.
The study of Langarpura, Neelidandi rock falls and Panjgran slump and rock fall are the first
documented examples of reactivated large scale mass movements on the hanging wall block of the
reactivated Muzaffarabad Fault and close to the epicentral region. Very few studies are available
on the description and analysis of factors which favour the reactivation of large scale mass
movements in the tectonically active region of the NW Himalayas. The result shows that the mass
movements occurred due to the earthquake reactivation of pre-existing mass movements. The
Langarpura and Neelidandi rock falls followed the steep scarp surfaces dipping southwest and
northwest respectively. The Langarpura rock fall reactivated southwest due to active southwest-
directed Himalayan thrusting. However, the Neelidandi rock fall reactivated due to northwest-
southeast active Himalayan extension perpendicular to the northeast-southwest active Himalayan
compression. The study indicates that these mass movements are the result of earthquake ground
motion, position on undercut slope causing over steepened slopes and preexisting reactivated
landslide bodies.
The reactivation of these mass movements on the hanging wall block of the Muzaffarabad Fault
suggests that the shaking level of ground motion was very high (Fig. 5.38; Intensity XI). The size,
texture, lithology, pre-existing scarp surfaces and morphology of older mass movement favored in
trigging these rock falls during the earthquake.
However, the study of the Panjgran slump and rock fall indicates that, it was an old mass
movement, which was reactivated during the earthquake 2005, followed the pre earthquake
Chapter 5: Mass movements triggered by the Kashmir earthquake 2005
107
escarpment failure. The slump and rock fall combines different types of movements. In the slump
zone the movement of material is dominantly rotational. The slump zone was destabilized due to
the erosion of Neelum river and during the reconstruction of the Neelum road. However, the rock
fall occurred at the top of the Panjkot ridge because the scarp area and the slide surface below the
scarp have been oversteppened and the bed rock of the Miocene Murree Formation is highly
fractured and sheared due to the ground shaking and close to the earthquake epicenter. The
analysis indicates that the mass movement is the result of preexisting slump on over steepened
slope undercut by the Neelum river, triggered by the Kashmir earthquake 2005.
Chapter 6: Statistical analysis of the mass movement distribution
108
Statistical analysis of the mass movement distribution triggered
by the Kashmir earthquake 2005
6.1. Introduction
The statistical analysis of mass movement distribution triggered by earthquakes has been
performed after many earthquakes in the world. Most studies investigated the general correlation
of mass movement distribution with causal factors such as earthquake source, ground motion,
shaking magnitude, terrain factors, and geological conditions (Jibson and Keefer, 1989; Keefer,
2000; Khazai and Sitar, 2003; Sato et al., 2007; Wang et al., 2007; Champati Ray et el., 2009; Qi
et al., 2010; Gorum T. et al., 2011). These studies provided valuable information in terms of
statistical distribution and characteristics of earthquake triggered mass movements, and have great
importance to understand the relationship between the distribution patterns of earthquake induced
mass movements and causal factors.
Kashmir earthquake 2005 generated more than 2,400 mass movements throughout the region in an
area of more than 7,500 km2, in the northern part of Pakistan (Sato et al., 2007; Owen et al., 2008).
Among them, 1,460 mass movements distributed in an area of approximately 1,299 km2, were
identified for statistical analysis of mass movement distribution, near the vicinity of the
Muzaffarabad city, the Jhelum valley and the Neelum valley areas. The area for analysis was
selected because it lies along the reactivated Muzaffarabad Fault and close to the epicentral region,
where the mass movement concentration was highest as compared to the other part of the affected
region. SPOT satellite imageries were used for the interpretation of the mass movement locations,
and field investigation was carried out in the specific zone along the rupture of the Muzaffarabad
Fault and the epicentral area. Therefore, a mass movement distribution map was prepared for this
study. The distribution of the events using SPOT satellite imageries (circular symbols) and ground
based field investigations (triangular symbols) are shown in Fig. 6.1.
6.2. Methodology
The analysis was carried out for all types of the mass movements induced by the Kashmir
earthquake 2005. The present procedure of analysis is similar to those described by Keefer (2000),
Khazai and Sitar (2003), Wang et al., (2007) and Qi et al., (2010) to investigate the mass
movement distribution induced by earthquakes in different parts of the world. ArcGIS 9.3 was
used to analyze the general correlation of mass movement distribution with causal factors (distance
from the earthquake source, topographic parameters and geological conditions).
Chapter 6: Statistical analysis of the mass movement distribution
109
Fig. 6.1 Mass movement distribution map of the Kashmir earthquake 2005 in the vicinity of
Muzaffarabad, Jhelum and Neelum valleys. The investigated area is marked by a rectangular
polygon.
Mass movement distances were calculated using two distance definitions; the epicenter and the
surface projection of the reactivated Muzaffarabad Fault. The position on the hanging wall and
foot wall block was determined. Topographic parameters such as slope angle, slope aspect and
slope elevation were calculated from the DEM and their dependency to the concentration of the
mass movements were determined. To understand the contribution of lithology, the analysis was
carried out using a geological map (Fig. 6.2) compiled after Wadia (1931), Calkins et al., (1975),
Chapter 6: Statistical analysis of the mass movement distribution
110
Baig and Lawrence (1987), Greco (1991), Hussain et al., (2004), Munir et al., (2006) and Kaneda
et al., (2008). The concentration of mass movements in different geological units was determined.
The dependency and effects of these different geological parameters on the concentration of the
mass movements were analyzed.
The index of the mass movement concentration for statistical analysis was defined to express the
influence of the mass movement occurrence according to Keefer (2000) and Wang et al., (2007),
which is calculated as number of mass movements per km2. Based on this criterion, the analysis
was performed for an area of 1,299 km2 which contained 1,460 mass movements.
Thus the average mass movement concentration of the study area was calculated as:
No. of mass movements: 1,460
Study area (km2): 1,299
Mass movement concentration (MCaverage) = 1,460 /1,299 km2
MC average
= 1.123 mass movements / km2
6.3. General mass movement distribution
In the study area, 1,460 mass movements triggered by the Kashmir earthquake 2005 covered 3.9 %
of the total surface area. As shown in Fig. 6.1, the mass movement distribution is higher in the
southeast of Muzaffarabad including area of Muzaffarabad city and Jhelum valley, followed by the
mass movement distribution in the area of the Neelum valley in the north and northeast of
Muzaffarabad city. However, the distribution of mass movements is very low in the south and
southwest direction of Muzaffarabad city. This asymmetric distribution of mass movements may
be caused by the increasing distance to the reactivated Muzaffarabad Fault and the epicentral area.
Preliminary field investigations showed that mass movements are very frequent along the local
streams, main roads and along the banks of the Neelum river and the Jhelum river. Most of the
mass movements were shallow failures. However, some large scale mass movements were also
observed. The volumes of these large scale mass movements were estimated more than 106 m3.
In general, a very high mass movement concentration was observed along the rupture zone of the
reactivated Muzaffarabad Fault, MBT and close to the epicentral area (Fig. 6.1). It can be observed
that mass movement concentration is mostly very high within a 20 km wide and 40 km long strip
along Muzaffarabad Fault, stretching in the northwest-southeast direction from Muzaffarabad to
Chikar. However, the distribution of the mass movements did not follow the same pattern and is
more widely dispersed around Muzaffarabad Fault on the hanging wall part. Likewise, the area
Chapter 6: Statistical analysis of the mass movement distribution
111
with highest mass movement concentration is restricted to only 10 km wide zone across the MBT
and epicenter region.
The concentration of mass movements is particularly dense along the hanging wall block of the
reactivated Muzaffarabad Fault as compared to the foot wall part. A large number of the mass
movements along the hanging wall block have very small in size and a shallow failure surface.
Nearly, all houses lying on the hanging wall block collapsed or sustained severe damage during
the earthquake, which suggests that the intensity of ground motion was very high in this area.
Additionally, the slopes of the hanging wall block of the reactivated Muzaffarabad Fault are very
steep, which is also factor for the triggering of so many mass movements during the earthquake in
this area.
The study area is mainly composed by weathered shales, siltstones, sandstones, claystones,
conglomerates, dolomites, limestones, marls, slates, phyllites and granitic or gneissic rocks
belonging to Precambrian, Cambrian, Carboniferous-Triassic, Tertiary and Quaternary rocks (Fig.
6.2). During the field survey it was observed, in the north of the study area, the mass movements
occurred in metasediments, graphitic schists, talc schists, marbles, coarse grained two-mica granite
gneisses, metacarbonates, metabasic volcanics, and quartzites of the Precambrian Tanol
Formation, the Cambrian Mansehra Granite and the Carboniferous-Triassic Panjal Formation.
Whereas, in the southeast of the area, most of the mass movements occurred in interbedded
sandstones, shales and claystones of the Miocene Murree Formation. Towards the west and
southwest of the study area the mass movements occurred along Muzaffarabad Fault and Jhelum
Fault in the Cambrian Muzaffarabad Formation and the Precambrian Hazara Formation. These
lithologies, played a vital role for the occurrence of the mass movements during the earthquake.
Chapter 6: Statistical analysis of the mass movement distribution
112
Fig. 6.2 Map showing the geological setting of the study area affected by the mass movements
(compiled after Wadia, 1931; Calkins et al., 1975; Baig and Lawrence, 1987; Greco, 1989;
Hussain et al., 2004, Munir et al., 2006 and Kaneda et al., 2008).
6.4. Mass movement concentration as function of distance to earthquake source
The statistical analysis of mass movement concentration as a function of distance to earthquake
source was made within the study area of 1,299 km2 (Fig. 6.1). Given the restriction of study area
being within the territory of effected part of AJK, this investigation was not carried out the whole
area, affected by the Kashmir earthquake 2005. Due to the geometrical effects, the area around the
earthquake source is unevenly distributed, that may misrepresent the overall results (Figs 6.3a and
Chapter 6: Statistical analysis of the mass movement distribution
113
6.3b). Considering the geometrical effects and the mutual effects of the Muzaffarabad Fault and
epicenter, a separate analysis was also made, involving an area of 10 km radius around the
epicenter and Muzaffarabad Fault (Figs. 6.3c and 6.3d).
Figs. 6.3 (a) Mass movement distribution around the epicenter within the whole study area, (b)
Mass movement distribution around the Muzaffarabad Fault within the whole study area (c) Mass
movement distribution involving an area of 10 km around epicenter, (d) Mass movement
distribution involving an area of 10 km around the rupture of the Muzaffarabad Fault.
Chapter 6: Statistical analysis of the mass movement distribution
114
6.4.1. Mass movement concentration in terms of distance from epicenter
The correlation between mass movement concentration and distance from the epicenter was
analyzed for all 1,460 mass movements, in the investigation area. The mass movement
concentration was obtained as a function of distance from the epicenter. Therefore, the distances
between the mass movement deposits and the epicenter were calculated. The mass movement
concentration was determined for a sequence of 1 km concentric bands extending (Buffer) up to 43
km outward from the epicenter in the southeast of the study area (Fig. 6.3a). The width (which
means the total distance from the epicenter to the mass movements of the concentric bands)
extending outward from the epicenter, was selected where the majority of the mass movements
was generated during the earthquake and provides enough details for the analysis. However, width
of the outer band towards north and northeast of Muzaffarabad district were truncated, as there
were less mass movements occurred during the earthquake. Towards northwest and southwest of
Muzaffarabad city, the bands were truncated due to the limitation of the study area.
Fig. 6.4 shows the variation of the mass movement concentration with the epicentral distance. In
general, the number of the mass movements fluctuates in the whole area. The highest mass
movement concentration (3.82 mass movements / km2) is reached in an area with up to 2 km
distance to the epicenter. Mass movement concentration values decrease from 3.82 mass
movements / km2 at a distance from 2-3 and 0.50 at a distance of 8-9 km to the epicenter, while the
mass movement concentration dramatically increases again at a distance of 19 km and 41 km away
from the epicenter, having mass movement concentration values of 2.41 and 2.75 mass movements
/ km2 respectively.
The analysis indicates that the mass movement concentration decreases with the increase in
distance up to 9 km from the epicenter (Figs. 6.4 and 6.5a). However, the concentration of the
mass movements doesn’t follow the same trend and gradually increases up to 19 km, beyond
which it again goes down and slightly fluctuates up to 39 km and at 41 km from the epicenter. This
rapid hike of mass movement concentration is apparently due to the proximity of the area to the
rupture of the Muzaffarabad Fault. There is clear evidence that these rises in mass movement
occurrence far from the epicenter are caused by the vicinity to the reactivated Muzaffarabad Fault
which was active during the 2005 earthquake. The rise in the mass movement concentration in 41
km distance to the epicenter is caused by the vicinity of the investigation area to Muzaffarabad
Fault and a geometrical effect, because only a little area directly in the vicinity to the
Muzaffarabad Fault is taken into account in this distance to the epicenter.
Chapter 6: Statistical analysis of the mass movement distribution
115
Fig. 6.4 Number, area of mass movement and mass movement concentration defined in terms of
distance from the epicenter.
The regression analysis for overall data shows a significant weak inverse relationship between the
distance from the epicenter and the mass movement concentration (R2 = 0.22, r = -0.46, p = 0.001;
Fig. 6.5a). This suggests that there is a decrease in mass movement concentration with increasing
distance from the epicenter. In addition, the standard error value of the mass movement
concentration compared to epicentral distance is considerably very high (Se= 11.23). This high
value of standard error may be due to the effect of the Muzaffarabad Fault rupture. The values fit
well empirically with a simple linear regression equation:
Mc = 1.9146 – 0.031De
Mc is the mass movement concentration and De is the epicentral distance in km.
However, there is a significant strong inverse correlation (R2 = 0.92, r = -0.94, p = <0.001; Se=
0.37) between the distance from epicenter and the concentration of mass movements, when an area
of 10 km away from the epicenter is put under regression analysis (Fig. 6.5b). The standard error
value of mass movement concentration with epicentral distance is 0.37. The data fit empirically
well with a linear regression equation.
Mc = 3.9973 - 0.3822De
Mc is the mass movement concentration and De is the epicentral distance in km.
Chapter 6: Statistical analysis of the mass movement distribution
116
Mc = 1.9146 - 0.031De
R2 = 0.22
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
0 5 10 15 20 25 30 35 40 45 50
Distance from epicenter, De (km)
Mas
s m
ov
emen
t co
nce
ntr
atio
n
Mc
(mas
s m
ov
emen
t /
km
2 )
Mc = 3.9973 - 0.3822De
R2 = 0.92
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
0 2 4 6 8 10
Distance from epicenter, De (km)
Mas
s m
ov
emen
t co
nce
ntr
atio
n
Mc
(mas
s m
ov
emen
t /
km
2 )
Figs. 6.5 (a) Relationship between the mass movement concentration and the distance from the
epicenter for over all data, (b) Relationship between the mass movement concentration and the
distance 10 km away from the epicenter.
(a)
(b)
Chapter 6: Statistical analysis of the mass movement distribution
117
This remarkable difference in the strength of inverse relationship between distance from epicenter
and mass movement concentration is likely due to the effect of Muzaffarabad Fault rupture. In
former instance, where whole data, adding the effect of Muzaffarabad Fault rupture, is involved,
the distance from epicenter and mass movement concentration shows a weak correlation and high
value of standard error. As the mass movement concentration gradually increased after 10 km
distance to epicenter due to the effect of Muzaffarabad Fault rupture that misrepresents the effect
of the epicenter on mass movement concentration. The analysis shows that the rupture of the
Muzaffarabad Fault strongly affected the concentration of the mass movements in terms of
distance from the epicenter for over all data. However, in a later case only data within 10 km
distance to the epicenter were taken into account. This analysis excludes the effect of
Muzaffarabad Fault rupture that increased the mass movement concentration away from the
epicenter. Therefore, the analysis indicates that in general the mass movement concentration
decreases with increasing the distance from the epicenter. The result shows that strong ground
motion caused the high concentration of mass movements near the epicentral region.
6.4.2. Mass movement concentration in terms of distance from Muzaffarabad Fault
The mass movement concentration in terms of distance from Muzaffarabad Fault rupture was
conducted for 1,460 mass movements. Therefore, the Muzaffarabad Fault rupture was used to
calculate the distance between the mass movement events and the fault plane rupture. One km
wide concentric bands have been constructed parallel to Muzaffarabad Fault extending outwards
(Buffer) from the hanging wall block of the Muzaffarabad Fault towards the northeast and
southeast, where the majority of the mass movements occurred during the earthquake (Fig. 6.3b).
The outer band towards south and southwest were truncated, as less mass movements occurred on
the foot wall block of the Muzaffarabad Fault. The largest distance between the mass movements
and Muzaffarabad Fault on the hanging wall side ranges up to 30 km, while on the foot wall side
the distance ranges up to 10 km, within the study area.
Fig. 6.6 shows the variation of the mass movement concentration occurring within one km
distance to the surface rupture of the reactivated Muzaffarabad Fault. As it can be seen from Fig.
6.6, the largest number of mass movements was found close to the fault and it is gradually
decreasing with the increase of the distance from the Muzaffarabad Fault rupture. While the
number of mass movements increases again in an area of 14-19 km distance to Muzaffarabad Fault
and then again decreases from 20-30 km distance to the fault rupture. Likewise, the values of mass
movement concentration are also very high, immediately adjacent to the Muzaffarabad Fault
rupture and decrease as the distance increases up to the distance of 11 km away from the fault. The
Chapter 6: Statistical analysis of the mass movement distribution
118
highest mass movement concentration value is found to be 2.34 at a distance of 1 km within the
surface rupture of the Muzaffarabad Fault. However, the concentration of the mass movements
gradually fluctuates from the distance of 15-19 km, in an area from the fault and having mass
movement concentration values of 0.48-2.09 mass movements / km2. Furthermore, the mass
movement concentration values again decrease to 2.00-0.18 mass movements / km2 in an area of
20-27 km distance to the Muzaffarabad Fault rupture.
The analysis shows that the concentration of the mass movements gradually decreases as the
distance increases up to 11 km away from the distance of the Muzaffarabad Fault. However, the
mass movement concentration values are also high at the distance of 15-24 km in an area from the
Muzaffarabad Fault. This indicates that the shaking motion is decreasing in general with
increasing distance from the Muzaffarabad Fault rupture, but at the distance of 15-24 km away
from the fault, the shaking motion is increasing, due to the effect of the epicentral area and
presence of the MBT in the northeast of Muzaffarabad. Although, the MBT was not reactivated
during the Kashmir earthquake 2005, but ground shaking level was very high due to close to the
epicentral region.
Fig. 6.6 Number, area of mass movement and mass movement concentration defined in terms of
distance from the Muzaffarabad Fault.
Chapter 6: Statistical analysis of the mass movement distribution
119
The linear regression analysis between the mass movement concentration and the distance from
the Muzaffarabad Fault rupture was statistically non significant for the overall data (R2 = 0.09, r =
-0.29; Fig. 6.7a). Furthermore, the standard error value (Se= 0.56) of mass movement
concentration with the rupture of the Muzaffarabad Fault is considerably high. The data can be
fitted with the linear regression equation:
Mc = 1.3315 – 0.0196Df
Where Mc is the mass movement concentration and Df is the distance from the Muzaffarabad
Fault in km.
However, there is the statistically significant inverse correlation between the mass movement
concentration and the distance from the Muzaffarabad Fault (R2 = 0.68, r = -0.82, p = 0.003; Se =
0.29, Fig. 6.7b), when the area of 10 km away from the Muzaffarabad Fault is put under regression
analysis. The values are well fitted with the linear regression equation:
Mc = 1.9133 - 0.1321Df
Where Mc is the mass movement concentration and Df is the distance from the Muzaffarabad
Fault in km.
Excluding the effect of the epicentral area and MBT, the analysis shows that there likely exists
moderate inverse correlation between the mass movement concentration and a certain distance.
Fig. 6.7a indicates that the relationship between the mass movement concentration and the distance
from the Muzaffarabad Fault for all data causes statistically non significant results. As the effect of
the epicentral region is involved, that falsifies the effect of Muzaffarabad Fault on mass movement
concentration. However, when considering the data within 10 km from the Muzaffarabad Fault as
shown in Fig. 6.7b. This indicates that the mass movement concentration decreasing with the
increase of the distance from the Muzaffarabad Fault. This reflects that mass movement
concentration around the high seismic zone, near the reactivated Muzaffarabad Fault may be the
result that rock mass in the adjacent area is highly fractured.
Chapter 6: Statistical analysis of the mass movement distribution
120
Mc = 1.3315 - 0.0196Df
R2 = 0.09
0.00
0.50
1.00
1.50
2.00
2.50
0 5 10 15 20 25 30
Distance from Fault, Df (km)
Mas
s m
ov
emen
t co
nce
ntr
atio
n
Mc
(mas
s m
ov
emen
t /
km
2 )
Mc = 1.9133 - 0.1321Df
R2 = 0.68
0.00
0.50
1.00
1.50
2.00
2.50
0 1 2 3 4 5 6 7 8 9 10
Distance from Fault, Df (km)
Mas
s m
ov
emen
t co
nce
ntr
atio
n
Mc
(mas
s m
ov
emen
t /
km
2 )
Figs. 6.7 (a) Relationship between the mass movement concentration and the distance from the
Muzaffarabad Fault for over all data, (b) Relationship between the mass movement concentration
and the distance 10 km away from the Muzaffarabad Fault.
(a)
(b)
Chapter 6: Statistical analysis of the mass movement distribution
121
6.4.3. Mass movement concentration in terms of distance from hanging wall and foot wall
blocks of Muzaffarabad Fault
Using the surface projection of the Muzaffarabad Fault, the mass movement concentration was
analyzed within a 14 km wide zone of 40 km length from Muzaffarabad to Chikar fault segment
along the hanging wall block and the foot wall block of the reactivated Muzaffarabad Fault. The
mass movement concentration was determined within one km concentric bands extending 7 km
away from the hanging wall block and foot wall block of the Muzaffarabad Fault (Fig. 6.8). The
width of the bands was truncated, as a very less number of mass movements occurred away from
this distance to the fault on the foot wall block and the given restriction of the study area. The
distances from the rupture of the reactivated Muzaffarabad Fault to the mass movements have
been calculated for the hanging wall block and the foot wall block.
The analysis shows that 40% of all mass movements occurred on the hanging wall side and 12 %
on that of the foot wall side of the Muzaffarabad Fault within the study area (Fig. 6.3b). Of these,
77 % of the mass movements occurred on the hanging wall block and 23 % mass movements in
the foot wall block within 7 km of the Muzaffarabad Fault which accounting for the area of 60 %
and 40 % respectively (Fig. 6.8).
Fig. 6.8 Mass movement distribution along hanging and foot wall blocks of the reactivated
Muzaffarabad Fault around 7 km distance, away from the Muzaffarabad Fault rupture.
Chapter 6: Statistical analysis of the mass movement distribution
122
Fig. 6.8 shows the variation of the mass movement concentration along the hanging wall block and
the foot wall block of the reactivated Muzaffarabad Fault. In general, the concentration of the mass
movements is related to the distance from the reactivated Muzaffarabad Fault rupture.
As seen in Fig. 6.9, the mass movement concentration has a maximum value of 3.40 mass
movements / km2 at a distance of 1 km on the hanging wall block of the Muzaffarabad Fault. The
mass movement concentration value drops to 0.87 mass movement / km2 at a distance of 7 km
from the fault rupture. While, the mass movement concentration values fluctuate from 1.54 to 0.39
mass movement / km2 in the distance of 7 km on the foot wall block of the Muzaffarabad Fault.
The analysis indicates that the concentration of the mass movements is high on the hanging wall
block of the Muzaffarabad Fault compared to the concentration of the mass movements on the foot
wall block. The values of the mass movement concentration decrease as the distance increases
from the fault source on both sides of the fault rupture.
Fig. 6.9 Number, area of mass movement and concentration of mass movement on the hanging
wall block (left) and foot wall block (right) of the Muzaffarabad Fault.
Chapter 6: Statistical analysis of the mass movement distribution
123
6.5. Mass movement concentration in terms of topographic parameters
An ASTER based DEM with 30 m x 30 m resolution of the study area was used for the analysis of
the dependency between mass movement concentration and topographic parameters (Figs. 6.10a,
6.10b, and 6.10c). According to the topographic characteristics, the analysis was performed among
the slope steepness, slope aspect and elevation categories and their relationship with the
concentration of the mass movement triggered by the Kashmir earthquake 2005.
For the analysis of topographic parameters, the study area is divided into seven categories based
on the slope angle. While, the slope aspect map is divided into eight slope aspect classes.
Likewise, the elevation in study area is divided into eight categories. 97% area is at the elevation
between 500–3,000 m asl, and only 3% is above 3,000 m asl for the total area (Tables 6.1, 6.2 and
6.3).
Fig. 6.10 (a) Slope angle distribution of the study area (b) slope aspect distribution of the study
area (c) DEM of the study area.
Chapter 6: Statistical analysis of the mass movement distribution
124
6.5.1. Mass movement concentration as function of slope steepness
In general, steep slopes are potentially endangered for mass movements triggered by earthquakes.
Within the study area the slope angle for the grid cells ranges from 0 - 70º and most of the mass
movements were found in grid cells with slopes ranging from 0 - 60º. For each 10º interval of
slope steepness, the mass movement concentration was calculated from the determination of the
grid cells and the number of the mass movements present within those grid cells. The percentage
of mass movement concentration was calculated as the percentage of the number of mass
movements divided by the percentage of mass movement area (Table 6.1).
Table 6.1 The relationship between mass movement concentration and slope steepness within the
study area of Kashmir earthquake 2005.
Slope
(degrees) No. of mass movements
Area (km
2)
Mass movement concentration
No. of mass movements
(%)
Area (%) Mass movement
concentration (%)
0-10 148 405 0.37 10.1 31.2 0.32 11-20 263 377 0.70 18.0 29.0 0.62
21-30 405 285 1.42 27.7 21.9 1.26
31-40 478 128 3.73 32.7 9.9 3.30
41-50 165 102 1.62 11.3 7.9 1.43
51-60 1 2 0.50 0.1 0.2 0.5
61-70 0 0.1 0 0 0 0
All 1,460 1,299 100.0 100.0
Fig. 6.11 and Table 6.1 show the relationship between occurrence (No. of mass movements) and
concentration of the mass movement and the slope angle categories. The mass movement
occurrence increases with increasing slope angle at categories of 31- 40º. It decreases with slope
angles bigger than 40°. Nearly 20 % of all mass movements occurred at slope angles less than 20
degrees accounted for the area of 51 %.
The slope categories of 0-10º contains larger area with less numbers of mass movements as
compared to the slope categories of 31-40º which contains majority of mass movements but less in
area. This shows that the higher slope angle affect the distribution of the mass movements.
The absence of mass movements in steeper slope categories is caused by the absence of debris in
these steep slopes, because most of the mass movements triggered by earthquake in this area are
caused by the mobilization of the debris cover, due to the earthquake ground motion.
The mass movement concentration also shows the same trend until highest concentration value of
up to 3.73 mass movements / km2 is reached in the class of 31-40° and then abruptly decrease to
1.61 mass movements / km2 at the category of 41-50º (Fig. 6.11).
Chapter 6: Statistical analysis of the mass movement distribution
125
Fig. 6.11 Number, area of mass movement and the relationship between mass movement
concentration and the slope gradient categories.
The analysis indicates that the mass movement concentration is highest at the slope angle of 31-
40º which accounted the area of 9.9 %. While, the second highest concentration is at the slope
angle of 41-50º covering an area of 7.9 %. The high concentration of the mass movements at slope
angle between 31-50º indicates that mass movements were strongly concentrated on steep slopes
close to the epicenter and Muzaffarabad Fault during the earthquake. The mass movement
concentration abruptly decreases at higher slope angles of 51-60º. This is may be due to the
absence of the debris at higher slope angle and the size of the grid cells used in the analysis.
6.5.2. Mass movement concentration as a function of slope aspect
Slope aspect has effects on the mass movements due to influencing the several factors like
insolation, weathered condition, land cover and soil condition (Kamp et al., 2008). The DEM
analysis of the study area reveals that the dominating mass movement directions are southwest,
south and southeast. The mass movements also facing slopes of other aspect as well (Fig. 6.10b,
Table 6.2).
Chapter 6: Statistical analysis of the mass movement distribution
126
Fig. 6.12 and Table 6.2 show the relationship between the occurrence (No. of mass movements)
and the concentration of the mass movement and eight slope aspect. The analysis indicates that,
nearly 57 % of all mass movements occurred on slopes facing in southeast, south, and southwest
directions which accounted an area of approximately 38 % of the study area. Followed by, 24 % of
mass movements facing in north, northeast and east directions covering approximately 31 % of the
area. The other 19 % of mass movements occurred in west and northwest direction accounting for
31 % of the area. This shows that mass movements distributed in slope facing in southerly
directions covering the larger area as compared to the other slope aspect.
Table 6.2 The relationship between mass movement concentration and slope aspect within the
study area of Kashmir earthquake 2005.
Slope
Aspect
No. of mass movements
Area (km
2)
Mass movement concentration
No. of mass movements
(%)
Area (%) Mass movement concentration
(%)
North 105 127 0.83 7.2 9.8 0.7
Northeast 94 132 0.71 6.4 10.2 0.6
East 149 135 1.10 10.2 10.4 1.0
Southeast 252 146 1.73 17.3 11.2 1.5
South 323 173 1.87 22.1 13.3 1.7
Southwest 259 176 1.47 17.7 13.6 1.3
West 177 204 0.87 12.1 15.6 0.8
Northwest 101 206 0.49 6.9 15.9 0.4
All 1,460 1,299 100.0 100.0
The mass movement concentration values are also very high in the southerly directions, as
compared to the other directions (Fig. 6.12). This suggests that the preferred orientation of the
mass movement seems to be dominated on the slope facing the southerly directions. This may be
due to the geometrical conditions of the valleys in the working region which cause many slopes
oriented in this direction.
Chapter 6: Statistical analysis of the mass movement distribution
127
Fig. 6.12 Number, area of mass movement and the relationship of mass movement concentration
and slope aspect.
6.5.3. Mass movement concentration as function of elevation
The study area is characterized by high steep rugged mountains and deep valleys. Mountains in the
northern part of the area are generally ranging 2,000-4,500 m asl. In the southern part the elevation
ranges from 500 m to 850 m asl. The area in the elevation between 500-1,500 m is mainly
composed of terraces and deep valleys. The area at elevation between 1,500 m to 2,000 m consists
of very steep slopes with debris cover.
Fig. 6.13 and Table 6.3 show the variation between the occurrence and concentration of the mass
movements and the elevation of the study area. Almost 74 % of all mass movements occurred at
elevations below 2,000 m asl, which accounted an area of 71 %. The 24.9 % of all mass
movements occurred at an elevation between 2,000–3,000 m asl covering an area of 25.6 %. Less
than 1 % of all mass movements occurred at elevations greater than 3,000 m asl accounting an area
of 3 %.
The analysis indicates that the maximum number of mass movements occurred on an elevation of
1,000-1,500 m (33.90%), 500-1000 m (23.84%) and 1,500-2,000 (16.23%), which is accounting
for 74% of the study area.
Chapter 6: Statistical analysis of the mass movement distribution
128
Table 6.3 The relationship between mass movements concentration and elevation within the study
area of the Kashmir earthquake 2005.
Elevation (m asl) No. of Mass
movements
Area
(km2)
Mass movement
concentration
No. of Mass
movements (%)
Area (%) Mass movements
concentration (%)
500-1000 348 376 0.92 23.8 28.9 0.82
1,000-1,500 495 288 1.71 33.9 22.1 1.53
1,500-2,000 237 257 0.92 16.2 19.7 0.82
2,000-2,500 188 172 1.09 12.8 13.2 0.97
2,500-3,000 177 162 1.09 12.1 12.4 0.97
3,000-3,500 13 25 0.52 0.8 1.9 0.46
3,500-4,000 2 15 0.13 0.1 1.1 0.12
>4,000 0 4 0 0 0.3 0.00
All 1,460 1,299 100 100
The number of the mass movements decreases with increasing elevation from 2,000-4,500 m.
However, the concentration of the mass movement is high between the elevations of 1,000-1,500
m (Fig. 6.13). The mass movement concentration abruptly decreases at the higher elevations. This
may be due to the area of higher elevation lies with larger distance from the Muzaffarabad Fault
and epicentral area. Moreover, only a small area lies over the higher elevations for the analysis.
Fig. 6.13 Number, area of mass movement and the relationship between mass movement
concentration and elevation of the study area.
Chapter 6: Statistical analysis of the mass movement distribution
129
6.6. Distribution of mass movements as function of geological units
In general, the rock masses within the study area are very fragile, highly fractured, sheared and
jointed due to folding and faulting. The bed rock geology of the area of Muzaffarabad city, the
Jhelum valley and the Neelum valley produced non uniform resistance against ground shaking,
structural failure, hanging wall collapse and escarpment failure during the Kashmir earthquake
2005. This is due to its varied inherent characteristics of the rock properties. Therefore, the
geological susceptibility to mass movements shows a significant distribution within the geological
units (Fig. 6.14).
The number of mass movements increases in areas where the lithology is prone for mass
movements, like Murree Formation of the Miocene age (Fig. 6.15). Furthermore, the mass
movement concentration is high in areas which are relatively close to the fault with high relief and
steep gradient. Therefore, the mass movement distribution for each geological unit is underlain by
nine major rock units as shown in Table 6.4 and Figs. 6.14, 6.15 and 6.16.
Table 6.4, Figs. 6.14, 6.15 and 6.16 show the occurrence of mass movements (No. of mass
movements), mass movement area, and mass movement concentration of different geological units
within the study area. The analysis indicates that 71% of all mass movements occurred in Tertiary
sediments, which are prone to the mass movement occurrence in sloping areas in different parts of
the world. Although, these units affected by the mass movements covered 79% of the whole study
area. In contrast, 29% mass movements occurred in the Precambrian, the Cambrian and the
Carboniferous-Triassic, rocks which accounted 21% of the area. Among the Tertiary sedimentary
rocks, the Miocene Murree Formation generated 67.4 % mass movements which cover 75.4 % of
the total area. Followed by Paleocene-Eocene sequence, Kamlial Formation and Quaternary
sediments produced 1.0 %, 0.1 % and 2.5 % mass movements accounted an area of 0.9 %, 0.8 %
and 1.8 % respectively (Figs 6.14 and 6.15).
Moreover, the second highest mass movement occurrence, 10.9 % was found in the Muzaffarabad
Formation, which accounted only 2.4% of the area. Followed by, mass movement occurrence 10.3
% in Manshera Granite, which accounted 10.2 % of the area. While, the mass movements
occurrence in Panjal Formation, Tanol Formation and Hazara Formation ranged 3.8 %, 1.6 %, 2.5
% which accounted 2.5 %, 2.4 % and 3.6 % of the area respectively (Figs. 6.15, 6.16 and Table
6.4).
Chapter 6: Statistical analysis of the mass movement distribution
130
Table 6.4 Geological formations, lithological description, age, percentage of mass movement,
percentage of surface area for geological units and mass movement concentration.
Geological
Formation
Description of Lithology Age Percentage
of mass
movements
(%)
Percentage
of surface
area (%)
Mass movement
concentration (mass
movements / km²)
Quaternary Stream bed deposits and alluvium.
Holocene 2.5 1.8
1.50
Kamlial Formation
Sandstones, shales, claystones and minor intraformational conglomerates.
Late Miocene 0.1 0.8 0.20
Murree Formation
Interbedded sandstones, siltstones with shales and claystones.
Early
Miocene
67.4 75.4 1.00
Paleocene-Eocene (Hangu, Lochart, Patala, Margala, Chorgali and Kuldana Formations)
Nodular limestones, calcareous and carbonaceous shales, claystones and laterite.
Paleocene-
Eocene
1.0 0.9 1.25
Panjal Formation Metacarbonates, metasediments, metabasic volcanics, quartzite and graphitic schists.
Carboniferous-Triassic
3.8 2.5 1.72
Muzaffarabad Formation
Cherty and stromatolitic dolomites, cherty white and grey bands, limestones and black shales
Cambrian
10.9 2.4 5.13
Mansehra Granite Coarse grained two-mica granite gneiss.
Cambrian 10.3 10.2 1.14
Tanol Formation Pelitic and psammitic metasediments, subordinate minor graphitic schist, talc schist and marbles.
Precambrian 1.6 2.4 0.74
Hazara Formation Slate, phyllite and shales with minor limestones and graphitic layers.
Precambrian 2.5 3.6 0.70
Chapter 6: Statistical analysis of the mass movement distribution
131
Quaternary
2.5%
Kamlial Fm
0.1%
Murree Fm
67.4%
Paleocene-Eocene
Sequence
1.0%
Panjal Fm
3.8%
Muzaffarabad Fm
10.9%
Mansehra Granite
10.3%
Tanol Fm
1.6%
Hazara Fm
2.5%
Fig. 6.14 The percentage of mass movement occurrence in terms of geological units.
The analysis in terms of mass movement area shows that the occurrence (No. of mass movements)
of mass movements is greater in the early Miocene Murree Formation rather than in other
formations (Fig. 6.14). The Miocene Murree Formation, which is one of the major geological units
within the study area consisting primary of shales, siltstones, claystones and interbedded
sandstones. Mass movements occurred along the bedding parallel slip between alternating
competent sandstones and in competent shales, siltstones and claystones. A similar pattern was
found in the Loma Prieta earthquake 1989, in which the Purisime Formation consisting of an
interbedded sequence of sandstones, siltstones and shales produced more mass movements than
any other geological unit (Keefer, 2000; Khazai and Sitar, 2003).
Chapter 6: Statistical analysis of the mass movement distribution
132
0
10
20
30
40
50
60
70
80
Quat
erna
ry
Kam
lial F
m
Murr
ee F
m
Pal
eoce
ne-Eoce
ne S
equen
ce
Panja
l Fm
Muza
ffar
abad
Fm
Man
sehr
a Gra
nite
Tanol F
m
Haz
ara
Fm
Geological units
Ma
ss
mo
vem
en
t a
rea
(%
)
0
1
2
3
4
5
Ma
ss m
ov
em
en
t co
nce
ntr
ati
on
Mass movement area Mass movement concentration
Fig. 6.15 The percentage of mass movement area in terms of geological units.
In this study, the concentration of the mass movements is within two main concentration zones.
One is the area close to the epicenter and MBT in the Neelum valley. While, the other one is along
the reactivated Muzaffarabad Fault on the hanging wall block in Muzaffarabad city and the Jhelum
valley area (Fig. 6.2). It can be seen from Fig. 6.16 that the highest mass movement concentration
with a value of about 5.13 mass movements / km2 was found in Cambrian Muzaffarabad
Formation exposed along the hanging wall block of the reactivated Muzaffarabad Fault. Whereas,
the second highest mass movement concentration value of 1.72 mass movements / km2 in Panjal
Formation exposed along the MBT and close to the epicentral area. The Quaternary sediment has
the third highest mass movement concentration value of 1.50 mass movements / km2. In contrast,
the lowest and the second lowest mass movement concentration values were found in the Kamlial
Formation and Tanol Formation ranging 0.20 and 0.74 mass movements / km2 respectively (Fig.
6.16 and Table 6.4).
The mass movement concentration differs substantially among various geological units within the
study area. However, the concentration of mass movements is higher in Cambrian Muzaffarabad
Chapter 6: Statistical analysis of the mass movement distribution
133
Formation as compared to other formations (Fig. 6.16), which indicates that the Muzaffarabad
Formation is highly susceptible to mass movements. This may have been aggravated due to the
closeness to the active Muzaffarabad Fault, the occurrence of intensely fractured and broken
dolomite rocks and very high steep slopes around Muzaffarabad city.
Fig. 6.16 Number, area of mass movement and the slope failure in different rock types defined in
terms of mass movement concentration.
The highest mass movement concentration in Muzaffarabad Formation with a value of 5.13 mass
movements / km2 accounted only for 2.4 % of the study area. While, the Murree Formation with a
mass movement concentration value of 1.0 mass movements / km2 accounted for 75.4 % of the
area (Figs. 6.15 and 6.16). This indicates that the mass movement concentration has no obvious
relationship with the area affected by the mass movements. However, the values of mass
movement concentration diminishing in Tanol Formation and Kamlial Formation may be due to
the given restriction of the area for this analysis rather than the lithological factor and earthquake
source. Therefore, other factors such as lithology, ground motion, collapsed of hanging wall block,
highly fractured rocks and the slope steepness can caused the variation of the mass movement
concentration values in different geological units within the study area.
Chapter 6: Statistical analysis of the mass movement distribution
134
6.7. Discussion and conclusions
The statistical analysis used herein shows that the distribution of mass movements triggered by the
Kashmir earthquake 2005 generally decreases with increasing distance from the epicenter and the
reactivated Muzaffarabad Fault. Results of the analysis show that the concentration of the mass
movements was high close to the earthquake source. However, for the overall data there are
sudden increases in concentration values away from the epicenter and the Muzaffarabad Fault
rupture, which involves merely the effects of both the epicentral area and the reactivated
Muzaffarabad Fault each other. Considering the data 10 km away from the earthquake source for
analysis, there are inverse correlations between the mass movement concentration and distances
from the epicenter and reactivated Muzaffarabad Fault. The highest coefficient of determination
for the epicenter (R2 = 0.92) suggests that the ground shaking that caused the mass movements was
highly concentrated near the epicenter rather than being uniformly distributed along the reactivated
Muzaffarabad Fault (R2 = 0.68). A similar relation was found in the previous studies for the Loma
Prieta event 1989 (Keefer, 2000), Chi-Chi earthquake 1999 (Khazai and Sitar, 2003), Chuetsu
earthquake 2004 (Wang et al., 2007) and Wenchuan earthquake 2008 (Gorum T et al., 2011).
For Loma Prieta, California event Keefer (2000) found a strong correlation (R2 = 0.97) between
the mass movement concentration and the epicenter rather than along the fault rupture (R2 = 0.80).
His relation is based on 1,280 mass movements mapped in an area of 2,000 km2. However, he
analyzed the data where the mass movement concentration was highest and data was relatively
complete. Khazai and Sitar (2003) performed the similar analyses for the Chi–Chi, Taiwan event
based on 2507 mass movement in an area of approximately 14,000 km2. They observed that the
mass movement concentration gradually decreases away from the epicenter and the surface
projection of the fault. They reported that the values of mass movement concentration diminishing
beyond the 40 km distance from the epicenter and mass movement concentration shows much
gradual decrease with distance away from the fault than the Loma Prieta event. Furthermore,
Wang et al. (2007) showed similar results for the Chuestsu earthquake Niigata, Japan. Their
analysis was based on 1212 mass movements covering an area of 275 km2. They concluded that
the mass movement concentration decreases with the increase of distance from the earthquake.
While, the percentage of the area affected by the mass movement increases with distance from the
epicenter. Contrary to these studies, Gorum T. et al. (2011) did not find any correlation between
the mass movement concentration and distance from the epicenter. However, a strong correlation
existed between the mass movement concentration and the distance from the fault rupture (R2 =
0.99). When compared the Kashmir earthquake 2005 results with Loma Prieta earthquake 1989
(Keefer, 2000), Chi-Chi earthquake 1999 (Khazai and Sitar, 2003) and Wenchuan earthquake
Chapter 6: Statistical analysis of the mass movement distribution
135
2008, China (Qi et al., 2010), its shows significant similarities in the concentration of mass
movements, while, considering the geometrical effects and the mutual effects of the earthquake
source, and contrary from Wenchuan earthquake 2008. As, the mass movement concentration for
the Wenchuan earthquake is primary controlled by the fault rapture rather than the epicenter. This
is due to the most of the mass movement events occurred within 10 km range of the fault.
In terms of topographic parameters, slope angle, slope aspect and elevation were analyzed to
reveal their correlation with the distribution of mass movements in the study area. Most mass
movements occurred in grid cells with slope angles ranging 31º- 40º and having mass movement
concentration values of up to 3.73 mass movements / km2.
The distribution of the slope angles for mass movements triggered with other earthquakes showed
that 90% of the mass movements occurred on slope angles greater than 45º in the Chi-Chi
earthquake 1999, while 83% and 90 % of mass movement failures occurred on slopes with less
than 50º in the Northridge and Loma Prieta earthquakes respectively (Khazia and Sitar, 2003).
However, most of the mass movements triggered by the Kashmir earthquake 2005 occurred on
slope angles of less than 50º, which is in close agreement to the Northridge and Loma Prieta
earthquakes and contrary from the Chi-Chi earthquake 1999. This difference from the Chi–Chi
earthquake is due to the mass movements typically occurred on steep slopes of young mountain
ranges in Taiwan (Lin et al., 2000).
Slope aspects have also effects on the distribution of the mass movements as well. The preferred
orientations of mass movement distribution were towards southerly directions. It should be noted
that most of the mass movement directions are related to the southwest Himalayan direction in the
northeast of the Himalayan of Pakistan.
The mass movement distribution in different elevation categories shows that a large number of
mass movements occurred on elevation ranges from 1000 -1,500 m which is accounting for 22 %
of the study area. The less number of mass movements at higher elevations is due to the small area
and larger distance from the earthquake source.
The distribution of mass movements in terms of geological units was examined according to the
description of lithology. Following the lithological description, 67.4 % of the mass movements
occurred in the Miocene Murree Formation which accounts 75.4 % of the total study area. In
contrast, the remaining 35% of the mass movements occurred in eight other geological units
covering the 25% of the study area.
Like the Chi-Chi earthquake 1999 and Loma Prieta earthquake 1989, the Kashmir earthquake also
triggered more than 70% of the total mass movements within Tertiary rocks. The mass movement
concentration (5.48 mass movements / km2) is much higher in the carbonate rocks of the Cambrian
Chapter 6: Statistical analysis of the mass movement distribution
136
Muzaffarabad Formation as compared to other formations within the study area. This might be due
to the highly fractured dolomites on the hanging wall block along the brittle shear zone of the
reactivated Muzaffarabad Fault.
This study concludes that the mass movement distribution is mainly depending on the distance
from the earthquake source (epicenter and Muzaffarabad Fault) rather than the topographic
parameters and geological settings of the area.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
137
Empirical analysis of geometrical parameters of mass movements triggered by
the Kashmir earthquake 2005
7.1. Empirical models
Empirical models are those that are based on observations and field data rather than on theoretical
assumptions or physical principles. These models are relatively simple and easy to use as
compared to other methods. The information used for these models can be easily collected in the
field or obtained from the existing literature. Sometime these models are also known as statistical
models (Keylock and Domaas, 1999).
In general, mass movement events and travel distance have been analyzed, using the two main
basic empirical models, widely adopted in the existing literature; the “Fahrböschung angle” (Heim,
1932; Shreve, 1968; Scheidegger, 1973; Hsü, 1975; Corominas, 1996; Erismann and Abele, 2001)
and the “shadow angle” (Lied, 1977; Hungr and Evans, 1988; Evans and Hungr, 1993). These
models correlate physical properties of the mass movements and establish the relationship between
the characteristics of the failure region and the travel distance (Mc Dougall and Hungr, 2004).
Empirical models are generally applied for the understanding of the factors such as initiation,
travel distance, and deposit volume to assess the mass movement hazard for the potentially
affected area. Furthermore, these models are widely used for the prediction of the travel distance
behaviour for mass movements in mountainous regions (Heim, 1932, Evans and Hungr, 1993).
The results can be used to predict the travel distance of mass movements in the same area or in
another area with similar conditions (Evans and Hungr, 1993; Soeters and Van Westen, 1996).
However, the results are not applicable to explain the mechanism and behaviour of mass
movements only by taking the travel distance into account (Copons et al., 2009). Therefore,
empirical methods should be applied with some judgment and care.
7.2. Study background
Before the Kashmir earthquake 2005, the risk posed by mass movements was underestimated and
no systematic scientific investigations were carried out in the study area. Sporadic scientific
investigations, carried out soon after the earthquake focused mainly on the identification and
distribution of mass movements. Owen et al. (2008) identified 1,293 mass movements at 174
locations within an area of 750 km2, near Muzaffarbad and Balakot, and developed first inventory
by quantifying the types of failure. However, they did not collect the data on important parameters
such as volume, length, height, Fahrböschung angle, shadow angle, and talus slope angle needed
for empirical analysis.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
138
This study was mainly focused on the vicinity of Muzaffarabad city, Jhelum valley, and Neelum
valley areas, in the NW Himalayas of Pakistan (Fig. 7.1). The data used for the analysis were
collected for the events spatially distributed on the steep slopes, the hanging wall block of the
reactivated Muzaffarabad Fault, the MBT and close in the epicentral region. Among several types
of mass movements, the analysis was restricted to mountain fall, rock fall and debris fall events
that are the predominant types of mass movements in the affected area (Fig. 7.2).
Fig. 7.1 Location of the mass movement events (represented by triangles), identified for empirical
analysis of geometrical parameters of mass movements, triggered by the Kashmir earthquake 2005
in the vicinity of Muzaffarabad city, Jhelum valley and Neelum valley. The investigated area is
marked by the rectangular polygon.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
139
7.3. Data source and methodology
Systematic field investigations were carried out after the identification of the mass movements on
SPOT images. The field data and SPOT images were used to map the events. A total of 103 rock
fall events were surveyed directly in the field for this analysis (Table 7.1). A mass movement
inventory, containing information of all geometrical parameters such as the volume, length, height,
surface area, deposit area, Fahrböschung angle, shadow angle and talus slope angle, needed for
statistical analysis was prepared (Appendix II). This inventory, produced from extensive ground-
based field work and SPOT satellite images, shows the locations and types of 103 mass
movements ranging in volume from 0.002 to 98 million cubic meters that occurred throughout an
area of approximately 1,299 km2.
Photographs of each event with full description (Appendix III) that also describes the type of the
mass movement have been taken. The description of the photographs preferably includes some
basic information about the event. Additionally, longitudinal profiles (Appendix III) that describe
the geometrical parameters of the event were prepared. Length (travel distance) and height (height
of fall) is likely to be measured from the longitudinal profiles fairly accurately. In contrast, the
average thickness has been estimated roughly at the deposit volume only and is mainly based on
the field assessment. The volumes of the rock falls were estimated by multiplying the deposit area
with an estimated average thickness. The volumes of the rock fall events that occurred along the
Jhelum river and the Neelum river were comparatively smaller than the actual rock mass displaced
from the scarp. This difference in volume was due to the erosion of the deposit material by the
river, especially during the flooding period. For mass movements with only small volumes it is
sometimes difficult to estimate the accurate volume due to hindrances like dense vegetation and
subsequent transportation of the material. In these cases the minimum estimate for the volume was
taken into account. Fahrböschung angle, shadow angle and talus slope angle were measured
directly in the field as well as verified from the longitudinal profiles. The surface area of the mass
movement is likely to be calculated accurately by using ArcGIS after interpretation of the mass
movement on SPOT images.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
140
Fig. 7.2 Examples of rock falls triggered by the Kashmir earthquake 2005, considered for
empirical analysis (Table 7.1). a) Battalian rock fall in the Jhelum vally, b) Makri rock fall in the
Muzaffarabad city, c) Nauseri rock fall in the Neelum valley, d) Devlian rock fall in Neelum
valley.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
141
Table 7.1 Geometrical data of 103 mass movement events triggered by the Kashmir earthquake
2005, in the vicinity of Muzaffarabad city, the Jhelum valley and the Neelum valley, in northern
Pakistan.
Nr. Type L (m) H (m) D (m) V (10
6m
3) Tan α Tan ß Tan δ
1 MF 2,350 700 60 98.4 0.3 0.24 0.36
2 MF 486 355 15 3.105 0.72 0.62 0.67
3 MF 805 460 12 5.76 0.57 0.56 0.53
4 MF 950 600 20 6.75 0.7 - -
5 RF 650 410 5 0.25 0.62 0.53 0.64
6 RF 800 480 3 0.189 0.6 0.55 0.64
7 RF 392 292 6 0.156 ��75 0.62 0.62
8 RF 258 150 6 0.138 0.57 0.56 0.56
9 RF 248 187 4 0.128 0.75 - -
10 RF 625 450 5 0.125 0.72 0.42 -
11 RF 436 346 4 0.12 0.78 0.7 0.72
12 RF 446 290 5 0.12 0.65 0.48 0.6
13 RF 206 192 4 0.112 0.93 0.56 0.72
14 RF 266 260 6 0.108 0.96 0.73 0.73
15 RF 416 310 4 0.108 0.73 0.42 -
16 RF 350 282 3 0.105 0.81 0.62 0.67
17 RF 288 248 6 0.102 0.86 - -
18 RF 298 238 6 0.102 0.8 0.62 -
19 RF 406 306 4 0.100 0.75 - -
20 RF 250 228 10 0.100 0.9 0.72 0.72
21 DF 230 166 2 0.088 0.72 0.62 -
22 DF 570 525 2 0.080 0.93 - -
23 DF 360 244 5 0.075 0.67 0.48 -
24 DF 330 274 4 0.072 0.83 0.72 0.72
25 DF 364 232 3 0.069 0.64 0.46 0.53
26 DF 300 240 1 0.066 0.8 0.67 0.67
27 DF 520 332 4 0.06 0.64 0.4 0.48
28 DF 248 202 3 0.060 0.81 0.64 0.78
29 DF 230 186 6 0.060 0.8 0.44 0.44
30 DF 252 150 4 0.060 0.6 - -
31 DF 254 170 2 0.060 0.67 0.62 0.62
32 DF 162 104 5 0.050 0.64 0.3 -
33 DF 175 130 5 0.050 0.75 0.4 0.62
34 DF 366 290 4 0.044 0.78 0.57 0.64
Chapter 7: Empirical analysis of geometrical parameters of mass movements
142
35 DF 306 296 3 0.042 0.96 0.53 0.53
36 DF 258 180 4 0.040 0.7 0.48 0.53
37 DF 320 242 3 0.039 0.75 - -
38 DF 270 156 3 0.036 0.57 0.48 0.57
39 DF 256 166 2 0.036 0.64 0.57 0.67
40 DF 190 110 3 0.036 0.57 0.38 0.48
41 DF 420 350 5 0.035 0.83 - -
42 DF 415 315 5 0.035 0.75 0.53 0.53
43 DF 400 300 2 0.032 0.75 - -
44 DF 282 230 3 0.030 0.81 0.53 -
45 DF 325 293 5 0.030 0.9 0.57 0.57
46 DF 210 190 3 0.030 0.9 - -
47 DF 162 146 4 0.028 0.9 0.78 0.78
48 DF 298 222 2 0.026 0.74 0.62 0.62
49 DF 174 122 2 0.022 0.7 0.62 0.64
50 DF 398 300 2 0.022 0.75 - -
51 DF 270 232 2 0.018 0.86 0.67 0.78
52 DF 405 292 2 0.018 0.72 0.62 0.67
53 DF 332 206 3 0.018 0.62 0.38 0.38
54 DF 382 326 1 0.018 0.86 - -
55 DF 210 126 3 0.018 0.6 - -
56 DF 230 240 1 0.017 1.03 - -
57 DF 195 116 3 0.016 0.6 - -
58 DF 144 134 3 0.015 0.93 0.72 0.78
59 DF 294 206 3 0.014 0.7 0.44 0.56
60 DF 136 126 2 0.014 0.93 - -
61 DF 136 85 2 0.014 0.62 - -
62 DF 386 282 3 0.012 0.72 0.48 0.6
63 DF 160 144 3 0.012 0.9 0.62 0.67
64 DF 202 182 3 0.012 0.9 - -
65 DF 180 162 2 0.012 0.9 - -
66 DF 140 109 4 0.012 0.78 0.7 0.7
67 DF 446 332 2 0.012 0.75 0.32 -
68 DF 196 164 1 0.011 0.83 - -
69 DF 150 118 2 0.010 0.78 - -
70 DF 132 90 5 0.010 0.67 0.36 -
71 DF 110 86 2 0.010 0.78 - -
72 DF 208 194 1 0.010 0.93 - -
73 DF 218 240 3 0.009 1.11 - -
Chapter 7: Empirical analysis of geometrical parameters of mass movements
143
74 DF 218 250 2 0.009 1.15 - -
75 DF 198 143 3 0.009 0.72 - -
76 DF 200 166 2 0.009 0.83 0.62 -
77 DF 350 222 2 0.008 0.62 0.53 0.6
78 DF 86 80 3 0.008 0.93 0.57 -
79 DF 218 270 3 0.008 1.23 - -
80 DF 137 100 2 0.007 0.72 - -
81 DF 175 127 2 0.007 0.72 - -
82 DF 76 70 5 0.007 0.92 - -
83 DF 168 151 2 0.007 0.9 0.48 0.57
84 DF 75 70 4 0.007 0.93 - -
85 DF 128 106 2 0.006 0.83 - -
86 DF 126 109 4 0.006 0.86 - -
87 DF 354 230 1 0.005 0.64 - -
88 DF 185 172 2 0.005 0.93 0.4 -
89 DF 130 125 3 0.005 0.96 - -
90 DF 166 155 2 0.005 0.93 - -
91 DF 88 71 2 0.005 0.81 - -
92 DF 94 90 4 0.005 0.96 - -
93 DF 100 86 1 0.004 0.86 - -
94 DF 112 121 1 0.004 1.08 - -
95 DF 98 86 2 0.004 0.83 0.64 0.72
96 DF 126 86 3 0.003 0.68 - -
97 DF 124 104 2 0.003 0.86 0.72 0.72
98 DF 330 250 1 0.003 0.75 - -
99 DF 254 350 1 0.003 1.37 - -
100 DF 232 200 3 0.002 0.86 0.44 -
101 DF 110 163 3 0.002 1.48 - -
102 DF 120 160 2 0.002 1.33 - -
103 DF 178 110 2 0.002 0.62 0.55 0.72
MF: Mountain Fall, RF: Rock Fall, DF: Debris Fall, L: Travel distance, H: Height of fall,
D: Estimated depth, Tan α: Tangent of the Fahrböschung angle, Tan ß: Tangent of the shadow angle,
Tan σ: Tangent of the talus slope angle, V: Volume
Note: (-) indicate the values of Tan ß and Tan σ are missing because the values were not measured
in field due to very steep narrow valleys and human modification had altered the area during the road
construction after the earthquake.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
144
7.4. Types of considered mass movements
The existing classifications of mass movements are based on different features like: process, type
of movement and activity, rate of movement, material involved, geometry, and morphology
(Varnes, 1978; Cruden and Varnes, 1996; Hungr et al., 2001).
A classification based on a volumetric nomenclature (Varnes, 1978 and Fell, 1994) for rock falls
was followed in this case, because our primary concern of rock fall classification involves travel
distance and Fahrböschung angle to explain the manner of progression of movement. The present
rock fall size classification based on a volumetric nomenclature has been used for the empirical
analysis (Table 7.2).
Table 7.2 Rock fall classification based on a volumetric nomenclature (after Varnes, 1978 and
Fell, 1994).
Types of falls Size Description Volume (m3) Description
Debris falls Very small <102 A large block or more than one block which may fragment during travel
Small 102 -103
Medium 103 - 104
Rock falls Large 104 -106 Free falling of rock
blocks of different size
or detached from the
rock slope
Mountain falls Extremely large >106 Fall, slide or avalanche
which may travel a
considerable distance
In general, rock fall, debris fall and mountain fall are defined as a rapid movement of rocks
triggered by earthquake, heavy rain or gravity forces from steep slopes in mountain areas. These
phenomena are a subsection of more general mass movement terms which include falls, slides and
slumps in all types of the material from hard rock material to unconsolidated or poor cemented
materials (Varnes, 1978; Keefer, 1999).
The term rock fall is characterized by the failure of relatively steep rock slopes or cliffs along a
surface where little or no shear displacement takes place (Varnes, 1978). In general, rock falls
involve a direct downward movement and small to medium detachments (104 – 106 m3), although
Chapter 7: Empirical analysis of geometrical parameters of mass movements
145
there is no well defined volume limit (Evans and Hungr, 1993). The low magnitude rock fall (<
104 m3) termed as debris fall (Varnes, 1978), is distinguished from a large rock fall (< 106 m3).
Extremely large rock falls (> 106 m3) are termed as mountain fall. The process is called sturzstrom
(Heim, 1932) or rock avalanche (Varnes, 1978; Keefer, 1984, 1999).
In this study, a large number of mass movements having a volume less than one million cubic
meters and only few mass movements having a volume greater than one million cubic meters are
analysed. The biggest mass movement has a volume of about 98 million cubic meters. For the
simplicity, rock falls have been further classified based on volume (Varnes, 1978 and Fell 1994;
Table 7.3)
Table 7.3 Classification of fall-types based on volume and number of mass movements.
Types of
falls
Size
Description
Volume in
million (m3)
Number
of falls
Fahrböschung
angle
Shadow
angle
Talus
slope
angle
Travel
distance
range
(m)
Debris
falls
Very small 0.002-
0.005
11 32º-56º 24º-36º 36º 94-330
Small .005-0.050 59 32º-49º 18º-38º 21º-
38º
88-446
Medium .05-0.10 13 33º-43º 17º-36º 24º-
38º
162-570
Rock falls Large 0.1-0.25 16 31º-44º 23º-36º 27º-
36º
250-800
Mountain
falls
Extremely
large
>1 4 17º-36º 486-
2,350
7.5. Geometrical parameters considered for empirical analysis
In this study, the geometrical parameters considered for the empirical analysis of mass movements
triggered by the Kashmir earthquake 2005 are Fahrböschung angle (Heim, 1932), shadow angle
(Lied, 1977; Evans and Hungr, 1993) and talus slope angle (Evans and Hungr, 1993) as presented
in Fig. 7.3. The term L refers to the travel distance or corresponding horizontal distance which is
the horizontal projection of the line connected from the source point of the mass movement and
the farthest block. The term H is the height of fall or elevation difference between the highest point
and lowest point of the mass movement (Fig. 7.3). The ratio H/L is defined as the tangent of the
Chapter 7: Empirical analysis of geometrical parameters of mass movements
146
Fahrböschung angle, the termed H/L ratio, is the ratio between the height of the fall and the travel
distance. Various geometrical approaches are related to the mass movement failure as shown in
Fig. 7.3.
Fig. 7.3 Sketch of mass movement source point, falling mass and deposit. Definitions of
parameters used in the present analysis are explained in text. Sketch is modified from Evans and
Hungr (1993) and Copons et al., (2009).
7.5.1. Fahrböschung angle
An important empirical approach is the Fahrböschung angle (Heim, 1932) or travel angle. The
Fahrböschung angle is the line connecting from the highest point of the mass movement scarp to
the distal margin of the displaced mass (Fig. 7.3). Various scientists referred this line by several
other names: angle of the equivalent coefficient of friction (Shreve, 1968), travel angle (Hungr,
1990; Cruden and Varnes, 1996), reach angle (Corominas, 1996), and travel distance angle
(Hunter and Fell, 2003).
Initially, the Fahrböschung angle method was widely adopted only for rock avalanches
(Scheidegger, 1973; Erismann and Abele, 2001). Later, Corominas (1996), Devoli et al., (2009)
and Copons et al., (2009) used the same approach for all types of large and small scale mass
movement events. Scheidegger (1973) reported negative linear relationship between volume and
Fahrböschung angle of 33 large scale mass movements. He noted the Fahrböschung angle
decreases with the increase of the volume. Therefore, larger mass movements display lower angle
of reach as compared to smaller ones, and due to this reason they were considered to be more
Chapter 7: Empirical analysis of geometrical parameters of mass movements
147
mobile. Corominas (1996) analyzed the similar relationship for 204 small and large scale mass
movements. He established the logarithmic relationship between mass movement volume and
Fahrböschung angle and shows a continuous decrease in angle of reach with increasing volume.
A similar relationship between volume of mass movement and Fahrböschung angle has also been
proposed by several other authors in various studies (Voight et al., 1983; McEwen, 1989;
Corominas, 1996;; Erismann and Abele, 2001; Legors, 2002; Hunter and Fell, 2003; Rickenmann,
2005; Devoli et al., 2009 and Copons et al., 2009). These studies showed also that the
Fahrböschung angle decreases, when the volume of the mass movement increases (>1x 105m3),
while, a constant coefficient of friction (0.57-0.83) can be assumed for smaller volumes of mass
movements. However, the limit of the mass movement volume has been revised by Hsü (1975),
and fixed to 0.5 x 106 m3. The value of the coefficient of friction is about equal to 0.6 for all types
of mass movements with smaller volumes. Corominas et al., (1988) also reported relatively low
reach angles for some of the mass movements ranging from few hundred cubic meters to several
thousand cubic meters in volume.
In contrary, opposite conclusion has been derived by Skermer (1985). He found no relationship
between the Fahrböschung angle and the volume of the mass movement. He suggested that the
height of the fall corresponds to the larger mobility of the mass movement instead of
Fahrböschung angle. Li (1983), Nicoletti and Sorriso-Valvo (1991) also found a similar
relationship between the height of the fall and the travel distance, but they suggested this
correlation is not applicable because the height of fall is not known in advance.
Different theories have been proposed by previous researchers to explain the phenomena of long
travel distance for the large mass movements, but none of them is consistent apparently with each
other (Van Gassen and Cruden, 1989; Melosh, 1986). The volume dependence of the
Fahrböschung angle has been questioned by many researchers for large scale mass movements
(Hsü 1975, Hungr, 1990) and small scale mass movements (Hunter and Fell, 2003). These
researches show that there is still a lack of agreement between researchers, and opposite
conclusion has been derived from these simple relations.
7.5.2. Shadow angle
An alternative approach followed by Lied (1977) is the shadow angle. The term shadow angle is
the line between the farthest block of deposit and the apex of the talus slope (Fig. 7.3). Lied (1979)
and Evans & Hungr (1993) described that the kinetic energy by rock blocks during their fall along
the trajectories is largely lost in the first impact on the talus slope, but this assumption is only
possible for small rock falls (<105 m3).
Chapter 7: Empirical analysis of geometrical parameters of mass movements
148
Previous scientists suggested several values of shadow angle in the existing literature. Evans &
Hungr (1993) suggested a value of 27º.5 after analyzing 16 rock falls in British Columbia.
Whereas, Lied (1977) proposed that the value can range from 28º to 30º. Wieczorek et al., (1999,
2008) proposed a shadow angle of 22º for Yosemite valley rock falls.
Domaas (1994) reported a smallest value of shadow angle of about 17º, whereas, Holm and Jakob
(2009) proposed a minimum shadow angle of 21º below a talus slope with fine debris. However,
they suggested that minimum shadow angle values are not applicable to other areas with different
lithological characteristics.
Copons et al. (2009) reported shadow angle values of 27º for the Sola d’ Andorra slope in Spain.
They indicated an inverse correlation between the size of the small rock falls (<100 m3) and the
travel distance by means of shadow angle. However, their plotted data have a high degree of
scattering.
7.5.3. Talus slope angle
There is another approach followed by Lied (1977), and Evans & Hungr (1993) is talus slope
angle. They have used the talus slope angle for the analysis of the nature of rock fall deposition.
The talus slope angle is the line between the lowest point and the apex of the talus slope (Fig. 7.3).
Evans & Hungr (1993) concluded approximately 38º talus slope angle for fine talus deposit,
however, lower down the talus deposit angle ranges from 32º-38º.
7.6. Analysis of Kashmir earthquake 2005 mass movement data
In this study empirical approaches are used to investigate the relationship between the geometrical
parameters and the travel distance of the mass movement events. The effect of different parameters
such as mass movement volume, Fahrböschung angle, shadow angle, talus slope angle, height of
fall are investigated to find a relationship to the travel distance of the mass movements. In total
103 mass movement events ranging in volume from 0.002 to 98.0 million cubic meters were
investigated for this study. Most of the mass movements were smaller in size (< 106 m3). Big and
little mass movements are included in the analysis to observe the influence of the different
volumes. Different mechanisms of failure, types of movement, rock types, and geomorphological
conditions were considered for this analysis. The relationship between the mass movement travel
distance and different parameters (volume, Fahrböschung angle, shadow angle, talus slope angle,
height of fall and surface area) were analyzed and presented in the form of empirical relationships.
The data are then compared with those published in earlier studies.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
149
7.7. Results of analysis
7.7.1. Relationship between mass movement volume and Fahrböschung angle for all types of
mass movements
In order to determine the relationship between the Fahrböschung angle and the mass movement
volume, an analysis of 103 mass movement events, was taken into account. These mass movement
events include mountain fall, very large and large rock falls, medium, small and very small debris
falls ranging in volume from 0.002 to 98 million cubic meters (Tables 7.1 and 7.2). In this
analysis, a large number of events with volumes less than 0.1 million cubic meters has been
included (Table 7.3). Most of the events triggered by the Kashmir earthquake 2005 have smaller
volumes, which represents an important limitation for this analysis.
Fig. 7.4 shows the relationship between the mass movement volume and the tangent of the
Fahrböschung angle for all types of mass movement events. Volumes versus log tangent of the
Fahrböschung values were plotted and results are presented. The values of tan α and volumes
plotted in a log–log plot ranged from 0.30 to 1.48 (log tan α – 0.52 to 0.17) and 0.002 to 98 million
cubic meters (log V-2.80 to 1.99), respectively.
A significant number of mass movement events with volumes less than 0.1 x 106 possess higher
values of tan α from 0.57 to 1.48 (log tan α – 0.24 to 0. 17), which mostly occurred in
Muzaffarabad and the surrounding areas along the Neelum river and the Jhelum river. For
instance, the events number 99, 101 and 102 having volumes less than 0.003 x 106 m3 (log v -2.59,
-2.62, -2.80) plotted in the left upper part of the plot on Fig. 7.4. These events display extremely
higher values of the tan α 1.37, 1.48, and 1.33 (log tan α 0.14, 0.17, 0.12). In contrary mass
movements that occurred in tributaries plotted in the left lower part of the plot had display lower
values of tan α as compared to the mass movements occurred along rivers. This may indicate the
deep incision of the Neelum and Jhelum rivers as compared to their tributaries with higher slope
angles towards to the rivers.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
150
Log tan α= -0.066 log V - 0.210
R2 = 0.29
-0.60
-0.50
-0.40
-0.30
-0.20
-0.10
0.00
0.10
0.20
0.30
-3.00 -2.00 -1.00 0.00 1.00 2.00 3.00
Log mass movement volume (106 m
3)
Lo
g t
an
Fa
hrb
ösc
hu
ng
All types of mass movements
Fig. 7.4 Relationship between log tangent of the Fahrböschung and log mass movement volume
for all investigated rock and debris falls triggered by the Kashmir earthquake 2005.
Moreover, in the dataset it can be observed, that the mass movements with larger volumes show
also higher values of tan α in some cases. For example, event no 2, 3 and 4 having volumes of 3.1
x 106 m3, 5.7 x 1 06 m3 and 6.7 x 1 06 m3 (log v 0.49, 0.76, 0.83) possess tan α values of 0.72, 0.57,
0.70 (log tan α -0.14, -0.24, -0.15) respectively, which are tremendously high (Table 7.1; Fig. 7.4).
These higher values of tan α may also be explained by the increasing influence of other factors
than volume such as the failure in a very steep slope and that the material will continue to move
until it reaches in the eroding river. However, these mass movements are several rock faces at one
location and difficult to distinguish with each other, so these mass movements were considered a
single event for this analysis.
Likewise, it was observed that the small volume of the mass movements varies the Fahrböschung
angle as low as the large volume of the mass movement. For example, event 103 has a value of tan
α = 0.62 for a volume of 0.002 x 106 m3, which is similar to the tan α of the event 5 (tan α = 0.62)
for a volume of 0.25 x 106 m3 (Table 7.1).
With respect to the topographical constraint and lithostratigraphic characteristics of the area, the
mass movements moving down the steep paths along the Neelum and Jhelum rivers are consider
less mobile and achieve higher Fahrböschung angle values. However, the Hattian Bala rock
Chapter 7: Empirical analysis of geometrical parameters of mass movements
151
avalanche with volume of 98 million cubic meters was found to be very mobile and displayed a
low Fahrböschung angle value of about 17º (tan α = 0.30).
For the data set of 103 events, the following empirical relationship has been obtained by using a
regression equation:
Log tan α = -0.066 log V - 0.210
Where tan α is defined here as has been earlier described in Fig. 7.3, the volume V is expressed in
cubic meters. The equation has a coefficient of determination (R2 = 0.29) and a coefficient of
correlation (r = -0.54). The standard error of the regression equation is 0.07.
In this relationship as shown in Fig. 7.4, the distribution of the data is rather scattered, but
reasonable, because all mass movements are integrated. However, these mass movements have
different type and mechanisms that can influence the travel distance.
The analysis shows there is a statistically significant inverse correlation between the tangent of the
Fahrböschung values and of the mass movement volumes. The coefficient of determination (R2 =
0.29) shows a minor correlation, than indicated in the previously published relations.
The results obtained may indicate that the volume of the mass movement had not such a strong
influence on the travel distance of the mass movements triggered by the Kashmir earthquake 2005,
may be because of the very high relief of the area.
7.7.2. Relationship between the mass movement volume and Fahrböschung angle for
individual groups of mass movements
To understand the effect of volume on Fahrböschung angle for each individual groups of rock
falls, the dataset of 103 mass movement events was split into different individual groups of rock
falls to be analyzed separately. The whole dataset was split mainly into rock falls and debris falls
based on the volume of the mass movements (Varnes, 1978 and Fell, 1994; Table 7.2). Moreover,
rock falls were further classified into extremely large and large rock falls. While, debris falls were
classified into medium, small and very small debris falls.
The mountain falls group ranging the Fahrböschung angles values from 17º-36º having volume
>1 x 106 m3, whereas, rock falls group has Fahrböschung angles ranging from 31º -44º with
volumes > 0.1-0.25 x 106 m3, while, the debris falls group belongs to Fahrböschung angles ranging
from 32º -56º having volumes < 0.1 x 106 m3 (Fig. 7.5, Table 7.3) .
Chapter 7: Empirical analysis of geometrical parameters of mass movements
152
Fig. 7.5 Fahrböschung angles and travel distances of individual groups of rock and debris falls
triggered by the Kashmir earthquake 2005.
In this analysis, the group of extremely large rock falls contains four large scale mass movements
having volumes > 1 x 106 m3 including the event of Hattian Bala (see detail in section 5.6.2) with a
Fahrböschung angle of 17º. The other three very large rock falls (> 1 x 106 m3) were also
reactivated during the earthquake and occurred along the reactivated Muzaffarabad Fault and close
to the epicenter region (see detail in section 5.6.3-5). However, these rock falls display higher
Fahrböschung angle values ranging from 32º -36º.
The group of large rock falls (<1 x 106 m3) are mostly found along the Neelum river and the
Jhelum river and their tributaries. These large rock falls occurred in a variety of geological
settings, some are associated with the reactivated Muzaffarabad Fault. The steeper Fahrböschung
angle in the dataset belongs to the group of debris falls with smaller volumes (<1 x 104 m3; Fig.
7.5). These debris falls occurred mainly nearby rivers, streams, roads and on low altitude areas.
The tangent of the Fahrböschung values were plotted versus the volumes of each individual groups
of rock falls and debris falls in a log – log plot (Fig. 7.6). The coefficient of regression analysis for
each individual group and their statistical information are presented in Table 7.4. The regression
equations for each group are as follows.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
153
Extremely large rock falls log tan α = -0.257 log V -0.004
R2 = 0.93
Large rock falls log tan α = -0.405 log V -0.489
R2 = 0.45
Medium debris falls log tan α = 0.250 log V + 0.158
R2 = 0.10
Small debris falls log tan α = -0.081 log V -0.244
R2 = 0.09
Very small debris falls log tan α = -0.271 log V -0.720
R2 = 0.08
-0.60
-0.50
-0.40
-0.30
-0.20
-0.10
0.00
0.10
0.20
0.30
-3.00 -2.00 -1.00 0.00 1.00 2.00 3.00
Log mass movement volume (106 m
3)
Lo
g t
an
Fa
hrb
ösc
hu
ng
Extremely large rock falls
Large rock falls
Medium debris falls
Small debris falls
Very small debris falls
Fig. 7.6 Relationship between tangent of the Fahrböschung and volume for extremely large rock
falls, large rock falls, medium debris falls, small debris falls, and very small debris falls.
The analysis shows the linear trend and an existing correlation between the tangent of the
Fahrböschung values and the volume of the large groups of rock falls (Fig. 7.6; Table 7.4).
Unfortunately, only four large scale mass movements were included for the analysis of the group
of extremely large rock falls (> 1 X 106 m3). This represents an important limitation in statistical
analysis for the group of very large rock falls. However, 16 rock fall events represented for the
group of large rock falls shows a good correlation (Table 7.4).
Chapter 7: Empirical analysis of geometrical parameters of mass movements
154
On the other hand, there is no clear correlation between the tangent of the Fahrböschung angle and
the debris falls with smallest volumes (<1 x 104 m3; Table 7.4). In addition, debris falls with
smaller volume (less than 0.003 x 10 6 m3) show the highest value of Fahrböschung angle, above
45º (tan α of 1.0) Figs. 7.5 and 7.6. This indicates that the trend of the mobility increases with
increasing volume, which has been previously accepted by many researchers. However, the travel
distance of debris falls with small volumes (<1 x 104 m3 ) varies and also depends upon to other
characteristics such as slope characteristic (Corominas et al., 1990), downhill path (Nicoletti and
Sorriso-Valvo, 1991; Corominas, 1996) and disintegration of the failure debris (Corominas, 1996;
Okura et al., 2000; Erismann and Abele, 2001).
Table 7.4 Results of the linear regression analysis of tangent of the Fahrböschung versus the mass
movement volume of individual rock fall groups.
Mass
movement
types
N V A B r R2 Se Comments
All mass movements
103 0.002-98 -0.210 -0.066 -0.54 0.29 0.07 Minor correlation
Extremely large rock falls
4 1.3-98.0 -0.257 -0.004 -0.96 0.93 0.05 Strong correlation
Large rock falls
16 0.1-0.25 -0.489 -0.405 -0.67 0.45 0.05 Good correlation
Medium debris falls
13 0.05-0.1 0.158 0.250 -0.31 0.10 0.07 No correlation
Small debris falls
59 0.005-0.05
0.244 -0.081 0.30 0.09 0.07 No correlation
Very small debris falls
11 0.002-0.005
-0.720 -0.271 -0.29 0.07 0.12 No correlation
Note: N, number of mass movements; V= volume in million m3; A and B, value of
regression coefficients; r, coefficient of correlation; R2, coefficient of determination; Se,
standard error
7.7.3. Relationship between mass movement volume versus shadow angle and talus slope
angle
An empirical approach is an effective method for estimating the approximate travel distance from
rock-slope failures of a volume well below that of large rock avalanches (< 106 m3). For the mass
movements triggered by the Kashmir earthquake 2005, an empirical approach proposed by Evans
Chapter 7: Empirical analysis of geometrical parameters of mass movements
155
and Hungr (1993) has been used to establish the relationship between the travel distance and the
slope failure for mass movements having volumes less than 1 x 106 m3.
The mass movement data used for this relationship were obtained during the field survey. At most
locations in the field, large accumulations of deposits from rock falls made it easy to measure the
shadow angles and the talus slope angles. However, this determination was more difficult in the
very steep narrow Neelum valley and Jhelum valley. Moreover, after the Kashmir earthquake
2005, human modification had altered the areas during road construction. As a consequence, only
those mass movement events were considered for this empirical approach, where all geometrical
parameters were measured directly as well as accurately during the field survey. Therefore on 55
mass movement events the exact shadow angle and on 41 mass movement events the exact talus
slope angle were ascertained for this analysis (Table 7.1).
The overall mean shadow angle is 28.6º for the selected mass movement events. Whereas, overall
mean talus slope angle is 32.2º for the selected mass movement events. The individual values of
the shadow angles vary from 17°-38º and the values for the talus slope angles vary from 21°-38º
(Table 7.3). The smallest shadow angle was observed to be about 17º for event no 32. The highest
value was noted to be about 38º for events no 47 and 51(Appendix II).
Similarly, the smallest talus slope value was found to be about 21º for event no 53, whereas, the
higher value was observed with about 38º for events no 28, 47, 51 and 58 (Appendix II). Most of
the mass movement events display similar values of shadow angle and talus slope angle, as the
deposit of the talus slope was accumulated near the rivers or streams.
The regression analysis was performed for the relationship between the shadow angle, talus slope
angle and the volumes (Figs. 7.7 and 7.8). The results of the analysis do not show a clear
relationship (R2 = 0.006) between volume and the distance traveled by means of shadow angle and
talus slope angle. Statistically, there is insignificant correlation. Moreover, the overall data are
highly scattered and very irregularly distributed (Figs. 7.7 and 7.8). The scattering of the data is
probably caused due to mass movements moving down the steep paths. It is not possible to find a
clear dependency between the volume of these talus slope deposits and the travel distance by
means of shadow angle and talus slope angle for this data set.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
156
-0.55
-0.50
-0.45
-0.40
-0.35
-0.30
-0.25
-0.20
-0.15
-0.10
-3.00 -2.50 -2.00 -1.50 -1.00 -0.50
Log mass movement volume (106 m
3)
Lo
g t
an
sh
ad
ow
an
gle
All types of mass movements
Fig. 7.7 Relationship between log tangent of the shadow angle and volume of 55 selected events
for all types of rock and debris falls triggered by the Kashmir earthquake 2005.
-0.45
-0.40
-0.35
-0.30
-0.25
-0.20
-0.15
-0.10
-3.00 -2.50 -2.00 -1.50 -1.00 -0.50
Log mass movement volume (106 m
3)
Lo
g t
an
ta
lus
slo
pe a
ng
le
All types of mass movements
Fig. 7.8 Relationship between log of the tangent of the talus slope angle and volume of 41 selected
events for all types of rock and debris falls triggered by the Kashmir earthquake 2005.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
157
7.7.4. Relationship between the Fahrböschung angle and the height of fall
To consider the effect of height of fall on Fahrböschung angle, a plot of tangent of the
Fahrböschung values against the height of fall is presented in Fig. 7.9. The values of Fahrböschung
angle (tan α) and height of fall ranged from 0.3-1.48 and 70-700 m respectively. For the majority
of the falls the Fahrböschung angle (tan α) is ranging from 0.6-1.0 and the heights of fall are
ranging between 70-300 m. (Table 7.1). It can be observed that extremely large rock falls and large
rock falls display a range of tan α values from 0.3 to 0.64 having a height of fall between 150-700
m. Medium, small and very small debris falls show values of tan α from 0.64-1.48 having a height
of fall between 70-525 m.
Statistically, there is no clear relationship (R2 = 0.03) between tangent of the Fahrböschung values
and height of fall. This indicates that the height of fall has no control directly on the travel
distance. However, it was observed that events with large height of fall travel also long horizontal
distances, but not necessarily display a lower Fahrböschung angle value. For example, event 4, the
Panjgran slump and rock fall traveled a horizontal distance of about 950 m with a height of fall of
600 m and displayed a Fahrböschung angle value of about 35º. Therefore, probably the height of
fall may affect indirectly controlling parameters of travel distance.
0
100
200
300
400
500
600
700
800
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Tan Fahrböschung
Hei
gh
t o
f fa
ll (
m)
Extremely large rock falls
Large rock falls
Medium debris falls
Small debris falls
Very small debris falls
Fig. 7.9 Relationship between tangent of the Fahrböschung and the height of fall for all types of
mass movements triggered by the Kashmir earthquake 2005.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
158
7.7.5. Relationship between the height of fall (H) and the travel distance (L)
The analysis of 20 rock falls (extremely large and large) and 83 debris falls (medium, small and
very small) was done to determine the relationship between the height of the fall and the travel
distance (Table 7.1 and 7.3). Height of fall (H) versus travel distance (L) values was plotted on a
log-log plot (Figs. 7.10 and 7.11). The height of fall and travel distance values ranged from 150 m
to 700 m (log 2.18-log 2.85) and 206 m to 2350 m (log 2.31-log 3.37) for rock falls respectively,
whereas, the values for the debris falls ranged from 70 m to 525 m (log 1.85-log 2.72) and 75 m to
570 m (log 1.88-log 2.76) respectively. For the data set of 20 rock fall events (Fig. 10) the standard
least square regression analysis give the empirical relationship:
Log L = 1.400 log H – 0.848
The equation has a coefficient of determination (R2) is 0.88 and a coefficient of correlation (r) of
0.94, where L and H are in meters. The standard error (Se ) is 0.08.
Log L = 1.400 log H - 0.848
R2 = 0.88
2.00
2.20
2.40
2.60
2.80
3.00
3.20
3.40
3.60
2.00 2.20 2.40 2.60 2.80 3.00
Log height of fall (m)
Lo
g t
ravel
dis
tan
ce (
m)
All types of rock falls
Fig. 7.10 Relationship between the log height of the fall (H) and the log travel distance (L) for 20
all types of rock falls triggered by the Kashmir earthquake 2005.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
159
For the data set of 83 debris fall events (Fig. 7.11), a best fit regression equation is obtained to:
Log L = 0.953 log H + 0.195
The coefficient of determination (R2) is 0.83. The coefficient of correlation (r) is 0.91. The
standard error (Se) is 0.08.
Log L = 0.953 log H + 0.195
R2 = 0.83
1.50
2.00
2.50
3.00
1.50 2.00 2.50 3.00
Log height of fall (m)
Lo
g t
ra
vel
dis
tan
ce (
m)
All types of debris falls
Fig. 7.11 Relationship between the log height of fall (H) and the log travel distance (L) for 83 all
types of debris falls triggered by the Kashmir earthquake 2005.
An analysis of the data shows that the travel distance is proportional to the height of fall for both
types (rock falls and debris falls) of mass movements (Figs. 7.10 and 7.11). This indicates that
there is a linear trend between the height of the fall and travel distance, as shown by the significant
positive relationships between the height of fall and travel distance for rock falls (R2 = 0.88,) and
debris falls (R2 = 0.83)
Available data were examined to identify the important factors that could influence the correlation
between travel distance and height of fall. Events with similar height of fall or similar volumes
were compared. The events 21, 39 (Table 7.1) occurred at the same height of fall of 166 m, but the
travel distance was different for each event. Similar are the case for events no 40 and 103.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
160
In the other hand, the Fahrböschung angle is also considered to affect the travel distance for rock
falls and debris falls in this analysis (Fig. 7.12). It has been clearly observed that mass movements
with the same Fahrböschung angle have different travel distances like events 3, 38 and 40.
Additionally long travel distances have been observed, when the rock fall passage ended in a
channel like surface indicating that geomorphologic control may be necessary to full explain
excess travel distance phenomena. This was for example observed by the events 43, 62, 77 (Table
7.1). The investigation shows that beside the volume, several other factors such as height of fall,
length, and Fahrböschung angle may have an influence on the observed travel distances.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
0 500 1000 1500 2000 2500
Travel distance (m)
Ta
n F
ah
rb
ösc
hu
ng
All types of rock falls
All types of debris falls
Fig. 7.12 Relationship between tangent of the Fahrböschung and the travel distance (L) for all
types of rock falls and debris falls triggered by the Kashmir earthquake 2005.
7.7.6. Relationship between surface area and volume of mass movements
In general, the larger the surface area of a mass movement, it is also likely to have a larger volume.
Therefore, an analysis of 103 mass movement events was done to determine the relationship
between the volume and surface area. Fig. 7.13 shows the values of the different surface areas and
volumes plotted on a log-log graph ranging from 0.003 to 2.02 km2 (log -2.52 to 0.31) and from
0.002 to 98 million m3 (log -5.80 to -1.01) respectively. The best fit equation obtained from the
least square regression analysis is:
Chapter 7: Empirical analysis of geometrical parameters of mass movements
161
Log A = 0.584 log V + 1.0537
The coefficient of determination (R2) is 0.79. The area A is in km2 and the volume V in km3
respectively. The standard error of the estimate is 0.23.
An analysis of the data reveals that there is a significant positive correlation (R2 = 0.79) between
the surface area and the volume of the mass movements. This indicates that the volume of the
mass movements increases proportionally to the surface area. The high standard error may indicate
the significant difference in variables. A similar relationship was found by Dortch et al., (2008) for
large landslides in the Himalayan region.
Log A = 0.5845log V + 1.0537
R2 = 0.7899
-3.00
-2.50
-2.00
-1.50
-1.00
-0.50
0.00
0.50
1.00
-7.00 -6.00 -5.00 -4.00 -3.00 -2.00 -1.00 0.00
Log volume (km3)
Lo
g a
rea
(k
m2)
All types of mass
movements
Fig. 7.13 Relationship between log surface area and log mass movement volume for all types of
mass movements triggered by the Kashmir earthquake 2005.
7.8. Comparison between international published data and own results
The relationship between the mass movement volume and Fahrböschung angle was studied by
Scheidegger (1973). Erismann & Abele (2001) studied this relationship especially for large rock
falls and rock avalanches. Corominas (1996) studied this relationship for all types of events
ranging from volume 102 - 108 m3 in the Eastern Pyrenees, northern Spain. A more recent similar
study was done by Devoli et al., (2009) for mass movements in the Central American countries for
Chapter 7: Empirical analysis of geometrical parameters of mass movements
162
all types of the landslides in volcanic and non-volcanic environments. Furthermore, Copons et al.,
(2009) analyzed small rock falls by using empirical models in southern Spain.
In order to understand the consistency of Kashmir earthquake 2005 data set with those published
in earlier studies in the different parts of the world the data published in several references were
collected and plotted on a log – log graph in Figs. 7.14 and 7.15. The used data are from
Scheidegger, 1973; Corominas, 1996; Legros, 2002; Erismann & Abele, 2003; Okura et al., 2003;
Devoli et al., 2009 and Copons et al., 2009). In total 509 events of rock falls, debris falls, rock
avalanches, debris avalanches, landslides, and debris flow were considered to show the
relationship between volume and the Fahrböschung angle. The 103 events triggered by the
Kashmir earthquake 2005 were also included in this analysis. However, different mechanisms and
environments have to be considered for the different types of these mass movements.
Fig. 7.14 shows the relationship between tangent of the Fahrböschung and volume of the mass
movements only for rock falls and rock avalanches from previously published data (Scheidegger,
1973; Erismann and Abele, 2001; Copons et al., 2009) compared with the Kashmir earthquake
2005 data set. Despite the diversity of the data and the differing mechanisms the results show a
tendency of a decrease of the Fahrböschung angle (tan α) with the increase of the volume of the
mass movements. This shows that the volume of the mass movement has a strong effect on travel
distance (Fig. 7.14).
Fig. 7.14 show that most of the scattered data near the left top of the plot belongs to the small rock
falls of Copons et al., (2009). The representing points of the large rock avalanches (Scheidegger,
1973; Erismann and Abele, 2001) are concentrated in the lower right part of the plot. Due to the
larger volumes of these large rock avalanches possess low tan α values which significantly reduced
scatter of the plot. The Kashmir earthquake 2005 data set involves mass movements with different
volumes from small rock falls to large rock avalanches The representing points are concentrated in
the middle of the plot. This means mass movements with smaller volumes are represent in more
scattered point clouds on the top left area of the plot as compared to the mass movements with
high volumes represented in the lower right part of the plot.
The Kashmir earthquake data set shows the higher values of tangent of the Fahrböschung similar
to Copons et al., (2009) data set. While, the data set of Scheidegger (1973), Erismann and Abele
(2001) shows the lower values of tangent of the Fahrböschung. This is may be due to the smaller
volume rock falls of Kashmir earthquake data set which have a different dynamic motion than the
larger one (>106 m3), as already accepted by many researchers (Corominas, 1996; Erismann and
Abele, 2001).
Chapter 7: Empirical analysis of geometrical parameters of mass movements
163
The analysis of the Kashmir earthquake data set reveals that there is weak or minor correlation (R2
= 29) between volume of mass movement and Fahrböschung angle (tan α), while those plotted by
Scheidegger (1973) and Copons et al., (2009) having coefficients of correlation (R2) ranging from
0.73 to 0.49). However, there is not much difference of correlation (R2 = 0.35) for the data set of
Erismann & Abele (2001) for large scale mass movements. This indicates that not only the volume
of the mass movement is influencing the travel distance.
Fig. 7.14 Relationship between log tangent of the Fahrböschung and log mass movement volume.
Comparison between the data of Scheidegger (1973), Erismann & Abele (2001), Copons et al.,
(2009) and the dataset of Kashmir earthquake 2005.
Fig. 7.15 compares the values of tangent of the Fahrböschung angle (tan α) versus volume of the
mass movements from international data set (Scheidegger, 1973;; Erismann and Abele, 2001;
Legros, 2002; Okura et al., 2003; Corominas, 1996; Devoli et al., 2009; Copons et al., 2009) with
the Kashmir earthquake 2005 data set. The international data include all types of mass movements
such as rock falls, rock avalanches, debris avalanches and landslides with different causes of
failures. While, the Kashmir earthquake data set considers only rock falls triggered by the
earthquake.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
164
For mass movements triggered by the Kashmir earthquake 2005, the best fit equation obtained by
using least square regression analysis gives the empirical relationship Log tan α = -0.066 log V -
0.210 having a coefficient of determination (R2) of 0.29, whereas the best fit equation obtained for
Corominas data is tan α = -0.085 log V -0.047 with a coefficient of determination (R2) is 0.62. The
best fit equation obtained for all international data is log tan α = -0.091 log V -0.816 having a
coefficient of determination (R2) of 0.34
The analysis shows the linear negative correlation between the volumes of the mass movements
and the Fahrböschung angles (tan α) values for all data sets.
As shown in Fig. 7.15, the representing points of the Kashmir earthquake dataset show higher
Fahrböschung angles compared for example with the dataset of Corominas (1996). A possible
reason can be, that the seismic triggering affected the travel distance of the mass movements of
Kashmir earthquake.
Comparing the international data set with Kashmir earthquake 2005 data set, the international data
fit and Kashmir earthquake data fit show similar trends (Fig. 7.15). The international data set
included a large number of mass movement events with large volume but with different
mechanisms of failure. Therefore, comparison of data sets shows the high scattered values for
larger mass movements as compared to the smaller ones (Fig. 7.15). This is totally different to the
results shown in Fig. 7.14, when we compared only rock fall and rock avalanche data sets. This
means beside the volume of the mass movements there are several other factors such as types of
mass movements, material involved and geological conditions affecting the scattering of the data.
Based on the observation of the different data sets, it can be concluded that the empirical
relationship suggested in earlier studies is mainly consistent with the data set of Kashmir
earthquake 2005. However, the Kashmir earthquake data set displays higher values of
Fahrböschung angle as compared to previous published data. Only Hattian Bala rock avalanche
shows a lower Fahrböschung angle, which is consistent with the earlier studies dealing with large
volume rock falls and rock avalanches.
Chapter 7: Empirical analysis of geometrical parameters of mass movements
165
-3.00
-2.50
-2.00
-1.50
-1.00
-0.50
0.00
0.50
1.00
1.50
-10.00 -8.00 -6.00 -4.00 -2.00 0.00 2.00
Lo
g t
an
Fah
rbö
sch
un
g
Log mass movement Volume (km3)
International dataKashmir earthquake data
Corominas, 1996
International data fit
Kashmir earthquake data fit
Corominas, 1996 data fit
Fig. 7.15 Relationship between log tangent of the Fahrböschung and log mass movement volume.
Comparison between international data and own data set (Kashmir earthquake 2005 data).
Chapter: 8 Conclusions and recommendations
166
Conclusions and recommendations
8.1. Conclusions
The Kashmir earthquake 2005 triggered a number of mass movements resulting in a great damage
and fatality in the Kashmir region, the northern part of Pakistan. The present analysis shows that in
the future more investigations are demanded on earthquake triggered mass movements, their
mechanism, characterization and distribution in order to improve the hazard assessment and
reconstruction process. The present study deals with the mass movement distribution,
characterization, and behaviour in the vicinity of Muzaffarabad city, Jhelum valley and Neelum
valley. A number of conclusions drawn from this study are as follows:
• The mass movement concentration is mainly depending on the distance from the
earthquake source rather than on topographical parameters and on geological units.
• The highest mass movement concentration (3.73 mass movements / km2) was found at
slope angle ranging from 31 – 40º. Moreover, the mass movement concentration was high
(1.7 mass movement / km2) at elevations between 1000–1500 m. In addition, the preferred
orientations of mass movements were towards southerly directions.
• The distribution of mass movements varies among different geological units. However, the
mass movement concentration is much higher in Cambrian Muzaffarabad Formation (5.13
mass movements / km2). This might be due to the highly fractured dolomites on the
hanging wall block along the brittle shear zone of the reactivated Muzaffarabad Fault.
• The empirical analysis showed that the Fahrböschung angle depends to some extent on the
volume. However, other factors such as slope steepness, very high relief of the area and
geological conditions can also affect the mass movement travel distance.
• Mass movements with only small volumes have variable values of Fahrböschung angle.
• No clear correlation was found between the Fahrböschung angle and the volumes of very
small debris falls.
• A linear trend with strong correlation exists between the height of fall and travel distance
for rock falls (R2 = 0.88) and debris falls (R2 = 0.83).
• The height of fall does not affect the Fahrböschung angle values.
• The empirical analyses show no relationship between the volume of the mass movements
and the travel distance by means of shadow angle and talus slope angle.
• The comparisons of Kashmir earthquake data with previously published data of the other
parts of the world are mainly consistent.
Chapter: 8 Conclusions and recommendations
167
• In the case study of Hattian Bala rock avalanche, the analysis leads to the conclusion that
the geometry and failure mode of this mass movement were strongly controlled by
tectonics and lithology, bedding parallel slip, southeast plunging synclinal structure and
pre-existing mass movements.
• In the case study of Neelidandi and Langarpura rock falls which were reactivated on the
hanging wall block of the reactivated Muzaffarabad Fault, the pattern of the mass
movements suggests that the existence of previous mass movements favoured the
triggering during the earthquake.
• In the case study of reactivated Panjgran slump and rock fall the mass movement followed
the pre-existing escarpment failure. The mass movement is the result of a pre-existing
slump situated on an over steepened slope undercut by the Neelum river.
8.2. Recommendations
In the light of the above conclusions, the following recommendations are made:
1. The national and international scientific community working in Pakistan Administrated
Kashmir, investigating mass movements triggered by the Kashmir earthquake 2005 should
provide a reliable systematic digital mass movement data base including economic and
human losses.
2. Mass movement inventory maps of the entire part of Pakistan Administrated Kashmir
should be prepared on ground based field investigations.
3. A detailed geological, structural and geotechnical mapping around the epicenter region and
reactivated Muzaffarabad Fault should be prepared.
4. Susceptibility maps and hazard maps should be prepared for future land use planning.
5. Many mass movements caused the disruption of the infrastructure in the affected part of
Pakistan Administrated Kashmir. The Highway department of Azad Jammu and Kashmir
government should come forward with scientific and technical backup for mass movement
control and management.
6. The centralized government institution should be established to provide better coordination
among various national and international scientific organizations to avoid duplicate efforts
for the mass movement investigation.
7. The Institute of Geology, university of Azad Jammu and Kashmir should establish the
discipline of disaster and mass movement studies. The discipline should conduct the
Chapter: 8 Conclusions and recommendations
168
research to improve the performance of mass movement data analysis by using most
appropriate methods and numerical modeling.
8. Efforts should be made to raise the public awareness of earthquake and mass movement
dangers in the Kashmir region.
References
169
References
Armbruster, J., Seeber, L., Jacob, K. B. (1978): The northwestern termination of the Himalayan
mountain front: active tectonics from micro earthquakes. J. Geophys. Res. 83: p. 269-282
Asian Development Bank and World Bank, (2005): Preliminary Damage and Need Assessment
(Pakistan Earthquake 2005): http://www.adb.org/Documents/Reports/pakistan-damage-
needs-assessment. PDF: p.1-124
Ashraf, M., Khan, M. S., Khan, M. Y. (1989): Petrology and geochemistry of the Panjal volcanics
in Poonch, Muzaffarabad and Kaghan valley, unpublished report, Pakistan Science
Foundation, Islamabad: p.114
Avouac, J. P., Ayoub, F., Leprince, S., Konca, O., Helmberger, D. V. (2006): The 2005, Mw 7.6
Kashmir earthquake: Sub-pixel correlation of ASTER images and seismic waveforms
analysis. Earth Planet. Sci. Lett., 249: p. 514-528
Baig, M. S., Lawrence, R. D. (1987): Precambrian to Early Paleozoic orogenesis in the Himalaya.
Kashmir J. Geol., 5: p. 1-22
Baig, M. S., Snee, L. W. (1995): The evidence for Cambro-Ordovician orogeny in northwest
Himalayas Pakistan. Geol. Soci. Am. Abstr. Program, 27: p. 305
Baig, M. S. (2006): Active Faulting and Earthquake Deformation in Hazara-Kashmir Syntaxis,
Azad Kashmir, northwest Himalaya, Pakistan. In: Kausar AB, Karim T, Khna T (eds)
Extended abstracts, International Conference on 8 October 2005 Earthquake in Pakistan:
Its implications and hazard mitigation: Geological Survey of Pakistan, Islamabad: p. 27-28
Baig, M. S., Yeats, R. S., MonaLisa., Pervez, S., Masood, B., Sohail, M. A., Jadoon, I. A.
K., Kausar, A. B. (2008): Active deformation, fault segmentation, scarp morphology,
seismic hazard assessments and geo-hazards along Muzaffarabad Fault, Hazara Kashmir
Syntaxis NW Pakistan, In: Mona Lisa, Jan MQ, and Khan MA (eds.). External abstracts In:
International Seminar “Earthquake hazards Pakistan: Post–October 2005, Muzaffarabad
Earthquake scenario” Baragali, Pakistan: p. 8-10
Bendick, R., Bilham, R., Khan, M. A., Khan, S. F. (2007): Slip on an active wedge thrust from
geodetic observations of the 8 October 2005 Kashmir earthquake. Geology, 35: p. 267-270
References
170
Bossart, P., Dietrich, D., Greco, A., Ottiger, R., Ramsay, J.G. (1984): A new structural
interpretation of the Hazara-Kashmir Syntaxis, southern Himalayas, Pakistan. Kashmir J.
Geol., 2 (1): p.19-36.
Bossart, P., Dietrich, D., Greco, A., Ottiger, R., Ramsay, J. G. (1988): The tectonic structure of
Hazara Kashmir Syntaxis, southern Himalayas, Pakistan. Tectonics, 7: p. 273-297
Bossart, P., Ottiger, R., Heller, F. (1990): Rock magnetic properties and structural development in
the core of the Hazara Kashmir syntaxis, NE Pakistan. Tectonics, 9(1): doi:
10.1029/89TC00968. ISSN: 0278-7407.
Brunsden, D. (1984): Mudslides. In Slope Instability (eds D. Brunsden and D.B. Prior), Wiley,
Chichester: p. 363-418.
Bull, W. B., King, J., Kong, F. C., Moutoux, T., Phillips, W. M. (1994): Lichen dating of
coseismic landslide hazards in alpine mountains. Geomorphology, 10(1): p. 253-264
Burbank, D. W., Reynolds, R.G. H. (1984): Sequential late Cenozoic structural disruption of the
northern Himalayan foredeep. Nature, 311: p.114-118
Calkins, J. A., Offield, T. W., Abdullah, S. K. M., Ali, S. T. (1975): Geology of the southern
Himalaya in Hazara, Pakistan, and adjacent areas. US. Geol. Surv. Prof. Paper, 716-C: p.
29
Calkins, J. A., Matin, A. S. A., Hussain, A., Mughal, M. N., Haq, I., Latif, A. (2004): Geological
map of the Garhi Habibullah area, District Mansehra and parts of Muzaffarabad District,
AJK: Geological Survey of Pakistan, Geol. Map Series v. VI, No., 13, Sheet No. 43 F/7,
1:50,000
Chaudhry, M. N., Ghazanfar, M., Ashraf, M. (1983): A Plate Tectonic Model for northwest
Himalayas. Kashmir J. Geol. Vol. 1, No., 1: p. 109-112
Chaudhry, M. N., Ghazanfar, M. (1992): Some tectonostratigraphic observations on Northwest
Himalaya, Pakistan. Pak. J. Geol. Vol., 1, No., 2: p.1-19
Chaudhry, M. N., Ghazanfar, M., Qayyum, M. (1986): Metamorphism at the Indo-Pak plate
margin, Kaghan valley, District Mansehra, Pakistan, Geol. Bull. Punjab Univ., 21: p. 62 –
86
References
171
Chaudhry, M. N., Ghazanfar, M. (1990): Position of the Main Mantle Thrust in the tectonic
framework of western Himalaya, Tectonophysics, 174: p. 321 – 329
Champati Ray, P. K., Parvaiz, I., Jayangondaperumal, R., Thakur, V. C., Dadhwal, V. K., Bhat, F.
A. (2009): Analysis of seismicity-induced landslides due to the 8 October 2005 earthquake
in Kashmir Himalaya. Curr. Sci., Vol. 97, No., 12: p 1742-1751
Chigira, M., Wu, X., Inokuchi, T., Wang, G. (2010): Landslides induced by the 2008 Wenchuan
earthquake, Sichuan, China. Geomorphology, 118: p. 225-238
Cruden, D. M., Varnes, D. J. (1996): Landslide types and Processes, in Landslides Investigation
and Mitigation. A. K. Turner and R. L. Schuster. Washington, D.C, Transportation
Research Board National Research Council. Special Report, 247: p. 36 - 75.
Copons, R., Vilaplana, J. M., Linares, R. (2009): Rock fall travel distance analysis by using
empirical models (Sol`a d’Andorra la Vella, Central Pyrenees). Nat. Hazards Earth Syst.
Sci., 9: p. 2107-2118
Corominas, J., Penaranda, R., Baeza, C. (1988): Identificacion de factores que condicionan la
formacion de movimientos superficiales en los valles altos del Llobregat y Cardener. II
Simp. Taludes y Laderas Inestables: 195-207. Andorra la vella.
Corominas, J. (1996): The angle of reach as a mobility index for small and large landslides.
Can. Geotech. J., 33: p. 260-271
Coward, M. P., Windley, B. F., Broughton, R., Luff, I. W., Petterson, M. G., Pudsey, C., Rex, D.,
Khan, M. A. (1986): Collision tectonics in the NW Himalayas, in Coward, M. P., Ries, A.
C., eds., Collision tectonics: Geol. Soci. London, Special Publication, 19: p. 203–219
Coward, M. P., Butler, R. W. H., Chambers, A. F., Graham, R. H., Izatt, C. N., Khan, M. A.,
Knipe, R. J., Prior, D. J., Treloar, P. J., Williams, M. P. (1988): Folding and imbrication of
the Indian crust during Himalayan collision, in Shackleton, R. M., Dewey, J. F., and
Windley, B. F., eds., Tectonic evolution of the Himalayas and Tibet: London, The
Royal Society: p. 89–115
Crawford, A. R., Davies, R. G. (1975): Ages of Pre-Mesozoic formations of the Lessor
Himalayan, Hazara district, northern Pakistan: Geological magazine, Vol., 12: p. 509-514
References
172
Devoli, G., Blasio, F. V., Elverhøi, A., Høeg, K. (2008): Statistical analysis of landslide events in
Central America and their run-out distance. Geotech. Geol. Eng. 27, No., 1: p.23–42
DiPietro, J.A., Pogue, K.R. (2004): Tectonostratigraphic Subdivisions of the Himalaya: A View
from the West: Tectonics, Vol. 23, TC5001, doi: 10.1029/2003TC001554.
Domaas, U. (1994): Geometrical methods of calculating rock fall range, Norwegian Geotechnical
Institute, Report, 585910-1: p. 21
Dunning, S. A., Mitchell, W. A., Rosser, N. J., Petley, D. N. (2007): The Hattian Bala rock
avalanche and associated landslides triggered by the Kashmir earthquake of 8 October
2005. Eng. Geol., 93 (3-4): p. 130-144
Erismann, T., Abele, G. (2001): Dynamics of rockslides and rock falls, Springer, Berlin: p. 316
Evans, S.G., Hungr, O. (1993): The assessment of rock fall hazard at the base of talus slopes.
Can. Geotech. J., 30: p. 620-636
Fell, R. (1994): Landslide risk assessment and acceptable risk, Can. Geotech. J., 31 (2): p. 261-272
Farah, A., Lawrence, R. D., De Jong, K. A. (1984): An overview of the tectonics of Pakistan. In
Marine Geology and Oceanography of Arabian Sea and Coastal Pakistan. Van Nostrand
Rienhold: p. 161-176
Farooq, M. S. (1997): Geotechnical study of landslide problems along the Murree-Muzaffarabad
road, Pakistan. PhD thesis, university of the Punjab, Lahore. p. 320
Fontan, D., Schouppe, M. (1995): Geological Map of the Neelum Valley, Azad Kashmir (NE
Pakistan) 1:50,000. In Regional Geological Mapping and ore potential assessment in
Neelum Valley, Azad Kashmir (Pakistan) EC Contract CI1-0571-M (GDF) Final Report.
Fujiwara, S., Tobita, M., Sato, H. P., Ozawa, S., Une, H., Koarai, M., Nakai, H., Fujiwara, M.,
Yarai, H., Nishimura, T., Hayashi, F. (2006): Satellite data gives snapshot of the 2005
Pakistan earthquake. EOS Transactions 87, 73 and 77
Gahalaut, V. K. (2006): 2005 Kashmir earthquake: not a Kashmir Himalaya seismic gap event.
Curr. Soci., 90(4): p. 507–508
References
173
Gallousi, C. I., Koukouvelas, K. (2007): Quantifying geomorphic evolution of earthquake-
triggered landslides and their relation to active normal faults. An example from the Gulf of
Corinth, Greece, Tectonophysics, Vol. 440, Issues, 1-4: p 85-104
Gansser, A. (1964): Geology of the Himalayas. Wiley Interscience, New York: p. 289
Gansser, A. (1981): The geodynamic history of the Himalayan. In Zagros-Hindu Kush Himalaya
Geodynamic Evolution, Am. Geophys. Union Geodynamics Series, Vol., 3: p. 111-121
Gee, E.R., (1989): Overview of the geology and structure of the Salt Range, with observations on
related areas of northern Pakistan, in Malinconico, L.L., Jr., and Lillie, R.J., eds., Tectonics
of the western Himalayas: Geol. Soci. Am. Special Paper, 232: p. 95–111.
Ghazanfar, M., Chaudhry, M. N. (1986): Reporting MCT in Northwest Himalaya, Pakistan.
Geol. Bull. Punjab Univ., 21: p.10-18.
Gorum, T., Fan, X., van Westen, C. J., Huang, R. Q., Xu, Q., Tang, C., Wang, G. (2011):
Distribution pattern of earthquake induced landslides triggered by the 12 May 2008
Wenchuan earthquake. Geomorphology, Vol. 133, Issue, 3-4: p. 152-167
Greco, A. (1989): Tectonic and metamorphism of the western Himalayan Syntaxis area (Azad
Kashmir NE Pakistan) dissertation ETH Zurich, 8779: p. 1-113
Greco, A. (1991): Stratigraphy, metamorphism and tectonics of the Hazara-Kashmir Syntaxis area.
Kashmir J. Geol., 8 & 9: p. 39-66
GSP (Geological Survey of Pakistan), (2008): Geological map of the Hattian Bala area.
http://www.gsp.gov.pk
Harp, E. L., Wilson, R.C., Wieczorek, G. F. (1981): Landslides from the February 4, 197
Guatemala Earthquake, US. Geol. Surv. Prof. Paper, 1204-A.
Harp, E. L., Crone, A. J. (2006): Landslides triggered by the October 8, 2005, Pakistan earthquake
and associated landslide-dammed reservoirs. US. Geol. Surv. Open-File Report,: p. 1-13
Harp, E. L., Jibson, R. W. (1996): Landslides triggered by the 1994 Northridge, California,
earthquake. Bull. Seismol. Soc. Am., 86(1B): p.319–332
References
174
Heim, A. (1932): Bergsturz und Menschenleben, Beiblatt zur Viertel-jahrsschrift der
Naturforschenden Gesellschaft in Zurich, 77: p. 218
Heim, A., Gansser, A. (1939): Central Himalaya– Geological observations of Swiss expedition,
1936: Mémoire, Société Helvetique Science Naturelle, Vol., 73 : p. 1–245
Holm, K., Jakob, M. (2009): Long rockfall runout, Pascua Lama, Chile. Can. Geotech. J., 46: p.
225–230
Hungr, O., Evans, S. G. (1988): Engineering evaluation of fragmental rock fall hazards,
Proceedings 5th International Symposium on Landslides, Lausanne, Switzerland, 1: p.
685–690
Hungr, O. (1990): Mobility of rock avalanches. Report of the National Institute for Earth Science
and Disaster Prevention, 46: p. 11-20
Hungr, O., Evans, S. G., Bovis, M. J., Hutchinson, J. N. (2001): A review of the classification of
landslides of the flow type. Environ. Eng. Geosci., VII (3): p.221-238
Hung, J. J. (2000): Chi-Chi earthquake induced landslides in Taiwan, International workshop on
annual commemoration of Chi-Chi earthquake, Sept. 18-20, 2000. National Center for
Research on Earthquake Engineering, Taipei: p. 23–36
Hunter, G., Fell, R. (2003): Travel distance angle for “rapid” landslides in constructed and natural
soil slopes, Can. Geotech. J., 40: p. 1123–1141
Hussain, A., Yeats, R. S. (2006): The Balakot–Bagh fault that triggered the October 8 earthquake
and other active faults in the Himalayan foreland region, Pakistan. In: Kausar AB, Karim T,
Hussain, S. S., Arshad, M. K., Shahzad, A. M. (2003): Mountains of Pakistan: Protection,
Potential and Prospects. Islamabad: Global Change Impact Studies Centre: p. 133-232
Hussain, A., Iqbal, S., Nasir, S. (2004): Geological maps of the Garhi Habibullah and Nauseri
area, District Muzaffarabad, AJK: Geol. Surv. Pakistan, Preliminary Map Series Vol..
VI, No., 14, Sheet No. 43 F/7, 11, 1:50,000
References
175
Hsü, K. J., (1975): Catastrophic debris streams (Sturzstroms) generated by rock falls: Geol. Soci.
Am. Bull. Vol., 86: p. 129–140
Hutchinson, J. N. (1968): Mass movement. In R. W. Fairbridge (Ed.), Encyclopedia of
Geomorphology, New York, Reinhold Publishers: p. 668–695
Hutchinson, J. N. (1988): General report, morphological and geotechnical parameters of landslides
in relation to geology and hydrogeology. In C. Bonnard (Ed.), Fifth International
Symposium on Landslides, Volume 1, A. A. Balkema / Rotterdam/Brookfield: p. 3–136
Iqbal, S., Nasir, S., Hussain, A. (2004): Geological map of the Nauseri area, District Muzaffarabad,
AJK: Geol. Surv. Pakistan, Geological Map Series Vol. VI, No., 14, Sheet No. 43 F/11,
1:50,000
Jibson, R.W., Keefer, D. K. (1989): Statistical analysis of factors landslide distribution in the New
Madrid seismic zone; Tennessee and Kentucky. Eng. Geol., 27: p. 509–542
Jibson, R. W., Harp, E. L., Schulz, W., Keefer, D. K. (2006): Large rock avalanches triggered by
the M 7.9 Denali Fault, Alaska, earthquake of 3 November 2002. Eng. Geol., 83 (1-3):
144-160
JSCE (Japan Society of Civil Engineers), (2006): Quick Report of the JSCE Mission for
Geotechnical Survey along Jhelum and Kunhar Valleys (Ver. 1.1): p. 1-11
Kamp, U., Growley, B. J., Khattak, G. A, Owen, L. A. (2008): GIS-based landslide susceptibility
mapping for the 2005 Kashmir earthquake region. Geomorphology, 101 (4): p. 631-642
Kaneda H., Nakata, T., Tsutsumi, H., Kondo, H., Sugito, N., Awata, Y., Akhtar, S. S., Majid, A.,
Khattak, W., Awan, A. A., Yeats, R. S., Hussain, A., Ashraf, M., Wesnousky, S. G.,
Kausar, A. B. (2008): Surface rupture of the 2005 Kashmir, Pakistan Earthquake and its
active tectonic implications. Bull. of the Seismol. Soci. Am., 98: p. 521-557: DOI
10.1785/0120070073
Kausar, A. B. (2008): Landslide disaster problems in 2005 Kashmir earthquake hit areas.
International conference on Management of landslide Hazard in the Asia-Pacific Region,
11th-15th November 2008. The Japan landslide society, Japan.
References
176
Kazuo, K., Jorgen, J., Shigeki, T., Takaaki, I. (2009): Landslide Disaster Risk Reduction. K. Sassa,
P. Canuti(eds), Springer-Verlag Berlin Heidelberg 2009: p.159-171
Keefer, D. K. (1984): Landslides caused by earthquakes. Geol. Soci. Am. Bull., 95: p. 6-21
Keefer, D. K. (1999): Earthquake-induced landslides and their effects on alluvial fans: Journal of
Sedimentary Research, Vol. 69, No., 1: p. 84-104.
Keefer, D.K. (2000): Statistical analysis of an earthquake-induced landslide distribution—the
1989 Loma Prieta, California event. Eng. Geol., 58: p. 231–249
Keefer, D. K. (2002): Investigating landslides caused by earthquakes. A historic review. Survey of
Geophys., 23: p. 473-510
Keylock, C., Domass, U. (1999): Evaluation of topographic models of rock fall travel distance for
use in hazard applications. Arctic, Antarctic, and Alpine Research 31(3): p. 312-320
Khan, M. S. (1994): Petrology and geochemistry of the Panjal volcanics in the Azad Kashmir and
Kaghan areas of the NW Himalaya. PhD thesis, University of Punjab, Pakistan: p.270
Khazai, B., Sitar, N. (2003): Evaluation of factors controlling earthquake-induced landslides
caused by Chi-Chi earthquake and comparison with the Northridge and Loma Prieta
events. Eng. Geol., 71: p. 79-95
Kazmi, A. H., Jan, M. Q. (1997): Geology and tectonics of Pakistan. Graphic Publishers, Karachi,
Pakistan: p.557
Kumahara, Y., Nakata, T. (2006): Active faults in the epicentral area of the 2005 Pakistan
earthquake. Hiroshima University Research Center for Regional Geography Special
Publication 41. Hiroshima University Research Center for Regional Geography,
Hiroshima: p. 54
Land Use Planning and Development department Muzaffarabad, (2007): Location map of
Muzaffarabad District AJ&K, Muzaffarabad, Pakistan: Scale 1:50,000
Latif, M.A. (1970): Stratigraphy and micropaleontology of Galis group of Hazara, Pakistan: Geol.
Bull. Punjab Univ. Vol., 13: p. 1-64
References
177
Latif, M.A. (1974): A Cambrian age for the Abbottabad Group of Hazara, Pakistan. Geol. Bull.
Punjab Univ., 10: p. 1-20
Legros, F. (2002): The mobility of long-runout landslides, Eng. Geol., 63: p. 301–331
LeFort, P. (1975): Himalaya: The collided range, present knowledge of continental arc: American
J. Sci. Vol., 275-A: p. 1-44
LeFort, P., Debon, F., Sonet, J. (1980): The "Lesser Himalayan" cordierite granite belt, typology
and age of the pluton of Manserah, Pakistan: Geol. Bull. Univ. Peshawar, 13: p.51-62
LeFort. (1981): P. Le Fort, Manaslu leucogranite: a collision signature of the Himalaya. A model
for its genesis and emplacement. J. Geophys. Res. Red Ser., 86 B 11: p. 10545–10568
Li, T. (1983): A mathematical model for predicting the extent of a major rock fall. Zeitschrift für
Geomorphologie, 24: p. 473-482
Li, T., Wang, S. (1992): Landslide Hazards and their mitigation in China. Science Press, Beijing,
p. 84
Lied, K. (1977): Rockfall problems in Norway, in: Rockfall dynamics and protective work
effectiveness, Instituto Sperimentale Modelli e Structture (ISMES), Bergamo, Italy, 90: p.
51–53
Lin, M. L., Wang, K. L., Chen, C. T. (2000): Characteristics of the Slope Failure Caused by the
Chi-Chi Earthquake, International Workshop on Annual Commemoration of the Chi-Chi
Earthquake, September 18–20, 2000, Vol., 3: p. 199–209
Marks, P., Ali, C. M. (1962): The geology of Abbottabad Area with special reference to the Infra-
Trias. Geol. Bull. Punjab Univ., 2: p. 1-56
McDougall, S., Hungr, O. (2004): A model for the analysis of rapid landslide motion across three-
dimensional terrain. Can. Geotech. J., 41(6): p.1084–1097
McEwen, A. S. (1989): Mobility of large rock avalanches: Evidence from Valles Marineris, Mars:
Geology, Vol., 17: p. 1111–1114
Melosh, H. J. (1986): The physics of very large landslides. Acta Mechanica, 64: p. 89-99
References
178
Middlemiss, C. S. (1896): The geology f Hazara and Black Mountains: Geological Survey of
India, Memoir, Vol., 26: p. 302
Middlemiss, C.S. (1910): The Kangra Earthquake of 4th April 1905, Memoirs of the Geological
Survey of India, Vol., 38: p. 409.
Molnar, P., Tapponnier, P. (1975): Cenozoic tectonics of Asia: Effects of a continental collision.
Science, Vol., 189. p. 419-426
MonaLisa, Khawaja, A. A., Jan, M. Q., Yeats, R. S., Hussain, A., Khan, S. A. (2008): New data on
the Indus Kohistan Seismic Zone and its extension into the Hazara Kashmir Syntaxis, NW
Himalayas of Pakistan. J. Seismol.: p. 1-15
MonaLisa, Khwaja, A. A., Jan, M. Q. (2008): The 8 October 2005 Muzaffarabad earthquake:
Preliminary seismological investigations and probabilistic estimation of peak ground
accelerations. Curr. Sci. 94, No., 9: p. 1158-1166
Munir, H. M., Baig, M. S., Mirza, K. (2006): Upper Cretaceous of Hazara and Paleogene
Biostratigraphy of Azad Kashmir, North-West Himalayas, Pakistan. Geol. Bull. of Punjab
Univ., 40-41: p. 69-87
Munir, M. H., Baig, M.S. (2006): Paleogene biostratigraphy of Tattapani, KotliAzad
Kashmir, northwest sub-Himalayas, Pakistan: Jour. Him. Earth Sci., Vol., 39: p. 39-48
Nakata, T., Kumahara, Y. (2006): Active Faults of Pakistan with reference to the Active Faults in
the source area of the 2005 North Pakistan Earthquake: In Extended Abstracts, International
conference on 8th October 2005 earthquake in Pakistan, Its implications and Hazard
migration, 18-19 January 2006, Islamabad: p. 18-20
Nakata, T., Tsutsumi, H., Khan, S. H., Lawrence, R. D. (1991): Active faults of Pakistan: Map
sheets and inventories. Special publication 21, Hiroshima University, Hiroshima: p. 141
Ni, J., Ibenbrahim, A., Rocker, S. W. (1991): 3 dimensional velocity structure and hypocenters of
earthquakes beneath the Hazara arc, Pakistan: Geometry of the under thrusting Indian
plate. J. Geophys. Res. Vol., 96 (19): p. 865-877
Nicoletti, P.G., Sorriso-Valvo, M. (1991): Geomorphic controls of the shape and mobility of rock
avalanches. Geol. Soci. Am. Bull., 103: p. 1365-1373
References
179
Okura, Y., Kitahara, H., Sammori, T., Kawanami, A. (2000): The effects of rock fall volume on
runout distance, Eng. Geol., 58: p. 109–124
Okura, Y., Kitahara, H., Kawanami, A., Kurokawa, U. (2003): Topography and volume effects on
travel distance of surface failure. Eng. Geol., 67: p. 243–254.
Owen, L. A., Kamp, U., Khattak, G. A., Harp, E. L., Keefer, D. K., Bauer, M. A. (2008)
Landslides triggered by the 8 October 2005 Kashmir earthquake. Geomorphology, 94: p.1-
9
Peduzzi, P. (2010): Landslides and vegetation cover in the 2005 North Pakistan earthquake: a GIS
and statistical quantitative approach. Nat. Hazards Earth Syst. Sci., 10: p. 623–640
Petley, D., Dunning, S., Rosser, N., Kausar, A. B. (2006): Incipient Landslides in the Jhelum
Valley, Pakistan following the 8th October 2005 earthquake. Disaster Mitigation of Rock
Flows, Slope failures and Landslides by Universal Academy Press: p. 1-9
Planning and Development department AJK, (2010): Azad Jammu and Kashmir at glance.
http://ajk.gov.pk
Patriat, P., Achache, J. (1984): India-Eurasia collision chronology and its implications for crustal
shortening and driving mechanisms of plates: Nature, Vol., 311: p. 615–621
Powell, C. M. (1979): A speculative tectonic history of Pakistan and surrounding: Some
constraints from the Indian ocean. A. Farah, K. A. de Jong (eds.). Geodynamics of
Pakistan, Geological Survey of Pakistan, Quetta (1979): p. 5-24
Qi, S. W., Xu, Q., Lan, H. X., Zhang, B., Liu, J.Y. (2010). Spatial Distribution Analysis of
Landslides Triggered by 2008. 5. 12 Wenchuan Earthquake, China. Eng. Geol., 116: p. 95–
108.
Rao, N. P., Kumar, P., Kalpana, Tsukuda, T., Ramesh, D. S. (2006): The devastating
Muzaffarabad earthquake of 8 October 2005. New insight into Himalayan seismicity and
tectonics. Gondwana Research, 9 (4): p.365-378
Rickenmann, D. (2005): Run-out prediction methods. In: Jakob M, Hungr O (eds) Debris-flow
hazards and related phenomena. Springer Berlin Heidelberg, Praxis: p. 305–321
References
180
Rustam, M. K., Sabir, M. K., Umar, F. (2003). Study of shallow geological structures in the core
of Hazara Kashmir Syntaxis based on Residual Gravity data in Azad Jammu and Kashmir
Pakistan. Geol. Bull. Punjab Univ., 8: p. 35-42
Sato, P. H., Hasegawa, H., Fujiwara, S., Tobita, M., Koarai, M., Une, H., Iwahashi, J. (2007):
Interpretation of landslide distribution triggered by the 2005 northern Pakistan earthquake
using SPOT 5 imagery. Landslide, 4: p. 113-122
Seeber, L., Armbruster, J. (1979): Seismicity of the Hazara Arc in northern Pakistan: decollement
vs. basement faulting. In: Farah A, Dejong KA (eds) Geodynamics of Pakistan, Quetta: p.
131-142
Schneider, J. F. (2008): Seismically reactivated Hattian slide in Kashmir, Northern Pakistan. J. of
Seismol.: p. 1-12. DOI 10.1007/s10950-008-9103-5
Scheidegger, A. (1973): On the prediction of the reach and velocity of catastrophic landslides.
Rock Mechanics, 5: p. 231- 236.
Shreve, R. L. (1968): The Blackhawk landslide: Geol. Soc. Am., Special Paper 108: p. 47
Skermer, N. A. (1985): Discussion of paper “Nature and mechanics of the Mount St Helens
rockslides-avalanche of 18 May 1980” Geotechnique, 35: p. 357-362
Soeters, R., van Westen, C. J. (1996): Slope instability recognition, analysis and zonation, in:
Landslides. Investigation and Mitigation, edited by: Turner, A. K. and Schuster, R. L.,
Transportation Research Board, National Research Council, Washington, special report,
247: p. 129–177
Schuster, R. L., Fleming, R. W. (1986): Economic losses and fatalities due to landslides. Bulletin
of the Association of Engineering Geologists, 23 (1): p. 11-28
Saba, S, B., Meijde, M., Werff, H. (2010): Spatiotemporal landslide detection for the 2005
Kashmir earthquake region. Geomorphology, 124: p. 17-25
Thakur, V. C. (2006): Reassessment of earthquake hazard in the Himalaya and implications from
the 2004 Sumatra-Andaman earthquake. Curr. Sci., 90 (8): p. 1070-1072
References
181
Treloar, P.J., Coward, M. J., Harris B. W. (1992): Hymalayan- Tibetian analogies for the evolution
of the Zimbabwe craton and Limpopo belt: Precambrian Research, Vol., 55: p.571-587
Treloar, P. J., Izatt, L. N. (1993): Tectonics of the Himalayan collision between the Indian plate
and Afghan block; a synthesis. Geological society, London, Special publication 1993, Vol.,
74: p. 69-87.
Tahirkheli, R. A. K. (1979): Geology of Kohistan and adjoining Eurasian Indo-Pakistan
continents, Pakistan, Geol. Bull. Univ. Peshawar, 11: p. 1 – 30.
Tahirkheli, R. A. K. (1982): Geology of the Himalaya, Karakoram, Hindukush in Pakistan. Geol.
Bull. Univ. Peshawar: p. 1-20
USGS (United States Geological Survey), (2006): Magnitude 7.6 Pakistan earthquake 2005
summary. www.earthquake.usgs.gov.
Van Gassen, W., Cruden, D. (1989): Momentum transfer and friction in the debris rock
avalanches. Can. Geotech. J., 26: p. 623-628
Varnes, D.J. (1958): Landslide types and processes, chap. 3 of Landslides and engineering
practice: Natl. Research Council, Highway Research Board Spec. Report, 29: p. 20-47
Varnes, D. J. (1978): Slope movement type and processes. In: Schuster RL, Krizek RJ, (Eds),
Landslides: analysis and control. National Academy of Sciences, Transportation Research
Board Special report, Vol., 76: p. 12-33
Varnes, D. J ed (1984): Landslide hazard zonation: a review of principles and practice. UNESCO
Ser. Natural Hazards, 3: p. 63
Voight, B., Janda, R. J., Glicken, H., Douglas, P. M. (1983). Nature and mechanics of the Mount
St. Helens rockslide avalanche of May 18, 1980. Geotechnique, 33: p. 243-273
Wadia, D. N. (1928): North Punjab and Kashmir. Rec. Geol. Surv. India, 62: p. 152-156.
Wadia, D. N. (1931): The syntaxis of the north–west Himalaya-its rocks, tectonics, and orogeny.
Rec. Geol. Surv. India, 65: p. 189-220
Wang, H. B., Sassa, K., Xu, W.Y. (2007): Analysis of a spatial distribution of landslides triggered
References
182
by the 2004 Chuetsu earthquakes of Niigata Prefecture, Japan. Nat. Hazards, 41: p. 43–
60
Wieczorek, G. F., Morrissey, M. M., Iovine, G., Godt, J. (1999): Rock-fall potential in the
Yosemite Valley, California: U.S. Geological Survey Open File Report 99-578, 1 plate,
scale 1:12 000, 7, http://greenwood.cr.usgs.gov/pub/open-file-reports/
Williams, H. R., Stott, G.M., Heather, K.B., Muir, T.L., Sage, R. P. (1991): Wawa Subprovince;
in Geology of Ontario, Ontario Geological Survey, Special Vol. 4, Part, 1: p.585-542
Windley, B. F. (1983): Metamorphism and tectonics of the Himalaya. Jour. Geol. Soc. London.
Vol., 140: p. 849-865
Wang, H. B., Sassa, K. (2006): Rainfall-induced landslide hazard assessment using artificial neural
networks. Earth Surface Processes and Landforms, 31(2): p.235-247
Yeats, R. S., Lawrence, R. D. (1984): Tectonics of the Himalayan thrust belt in northern Pakistan.
In Marine Geology and Oceanography of Arabian Sea and Coastal Pakistan. Van Nostrand
Rienhold: p. 177-198
Yeats, R. S., Parsons, T., Hussain, A., Yagi, Y. (2006): Stress changes with the 8 October 2005
Kashmir Earthquake, Lessons for the Future. In: Kausar AB, Karim T, Khna T (eds)
Extended abstracts, International Conference on 8 October 2005 Earthquake in Pakistan:
Its implications and hazard mitigation: Geological Survey of Pakistan, Islamabad: 16-17
183
Appendices
Top Related