ASPECTS OF EXHUMATION OF ROCKS IN EXTENSIONAL AND COMPRESSIONAL MOUNTAIN BELTS. A COMPARATIVE STUDY

A thesis submitted to the Faculty of Natural Sciences, Karl-Franzens-Universität Graz (University of Graz), Austria, in partial fulfillment of the requirements for the degree of Doctor of Science

By

SYED ALI TURAB

Institute of Earth Sciences Karl-Franzens-Universität Graz (University of Graz) Austria (November 2016)

Acknowledgements

I would like to thank Kurt Stüwe for supervising this work and providing continuous support and freedom during these four years. Fin Stuart (SUERC, Glasgow) is also thanked for his help and guidance during lab analysis. David Chew and Nathan Cogné (TC, Dublin) are acknowledged for AFT and apatite U-Pb analysis and providing help in understanding and running thermal history modelling software, the QTQt. Luigia Di Nicola, Mahmoud Hassan, Gustav Hanke, Tamer Abu-Alam, Jamil Hassan, Gisela Domej, Angela Oswald, Sylvia Umschaden, Daniel Döpke, Katarzyna Luszczak, Christoph Pucher and Elena Sizova are also thanked for providing help at various stages. All thanks to the friends in the Saudi Geological Survey (SGS) Geologic Mapping Unit, headed by Khalid A. Kadi and the field group Saad M. Al Garni, Mubarak M. Al Nahdi and Abdullah Al Shammari are thanked for providing help in the field.

This work was financially supported by Higher Education Commission (HEC) of Pakistan and Austrian Agency for International Cooperation in Education and Research (OeAD- GmbH) at different levels. Partial support for the field work was provided by Heinrich-Jörg Stiftung funding scheme of University of Graz, Austria, National Centre of Excellence in Geology (NCEG), University of Peshawar, Pakistan, and Saudi Geological Survey (SGS).

At last but not the least, I thank my family and friends whose constant support helped me to achieve this goal.

i

Turab, S.A., 2016

Table of Contents

Acknowledgements ...... i Table of Contents ...... ii Preamble and Outline of the thesis ...... iv Abstract ...... vi Zusammenfassung ...... viii Chapter 1. New constraints on exhumation of the western Himalayan syntaxis. A low temperature thermochronometry of the Neelum River region, Pakistan ...... 1 ABSTRACT ...... 2 INTRODUCTION ...... 3 GEOLOGICAL BACKGROUND ...... 5 LOW TEMPERATURE THERMOCHRONOLOGY: SAMPLES AND TECHNIQUES ...... 8 THERMOCHRONOLOGICAL RESULTS ...... 10 MORPHOLOGICAL ANALYSIS ...... 16 DISCUSSION ...... 20 Tectonic Interpretation ...... 22 Cause of Exhumation ...... 25 CONCLUSIONS ...... 27 Chapter 2. Escarpment evolution at the Red Sea continental margin of southwestern Saudi Arabia ...... 29 ABSTRACT ...... 30 INTRODUCTION ...... 31 GEOLOGICAL BACKGROUND ...... 35 LOW TEMPERATURE THERMOCHRONOLOGY: SAMPLES AND TECHNIQUES ...... 37 THERMOCHRONOLOGICAL RESULTS ...... 38 Line 1 ...... 40 Line 2 ...... 40 Line 3 ...... 47 Line 4 ...... 48 MORPHOLOGICAL ANALYSIS ...... 48 DISCUSSION ...... 50 Total Amount and Timing of Uplift ...... 50 Escarpment Evolution ...... 55 Concept of Rifting/Active Vs Passive Rifting ...... 57 ii

Turab, S.A., 2016

CONCLUSIONS ...... 60 Chapter 3. Overall Conclusions ...... 61 Appendix A ...... 64 Appendix B ...... 69 References ...... 76

iii

Turab, S.A., 2016

Preamble and Outline of the thesis

Exhumation of rocks in orogenic belts may occur due to the three end member processes of (a) extension, (b) erosion and (c) selective exhumation in compressional tectonics due to ductile flow (Ring, et al., 1999). In extensional environments, normal faulting can directly be responsible for the exhumation of rocks. The classic example of this type of exhumation is the formation of metamorphic core complexes (Lister and Davis, 1989), where normal faults are the direct agent responsible for the exhumation of high grade metamorphic rocks from mid crustal levels to the surface. In the field, this process is evidenced by the presence of normal faults bounding the exhumed rocks. Conversely, erosion can also be directly responsible for the exhumation. Clearly, this process is only possible if there is topography, so that exhumation by erosion is usually directly correlated with mountain building. In compressional mountain belts, erosion is the typical exhuming process (e.g. England, 1981). Finally, selective exhumation by ductile flow is a process that allows spatially confined bodies of rocks to be squeezed to the surface, often in compressional environments. The exhumation of individual eclogite bodies in subduction zones is an example for this process (Chemenda, 1995).

Much debate has been given to the relative importance of these three processes in orogenic belts. One of the principal discriminating pieces of evidence is the distribution of exhumation of rocks in space and time and its relationship to active deformation features. In order to contribute to this debate, two of the globally most spectacular examples of mountain building in contrasting tectonic environments were chosen and the exhumation processes studied. In particular, I have chosen: (a) the western syntaxis region of the Himalayan orogeny around Nanga Parbat as a text-book example to study the exhumation processes in a mountain belt formed by collision of two plates and (b) the 3000 m high mountains of the western Saudi Arabian escarpment as an exemplary mountain belt formed due to a divergent plate margin, namely that between Africa and the Arabian sub-continent. In my study, I have focused on the exhumation of rocks in the uppermost crust, a process that can be studied with the aid of low- temperature thermochronology. Accordingly, the thesis is structured in the form of three chapters and two appendices.

In Chapter 1, a multi-method approach is used to study the exhumation of rocks in northwest Himalayan syntaxis region. So the geomorphic analysis of the region are combined with results of multiple thermochronometric methods including apatite U-Pb and fission track as

iv

Turab, S.A., 2016 well as (U-Th-[Sm])/He dating of apatite and zircon. This chapter is submitted to Lithosphere and is currently under review:

 Turab, S.A., Stüwe, K., Stuart, F.M., Chew, D.M., and Cogné, N., 2016, New constraints on exhumation of the western Himalayan syntaxis. A low temperature thermochronometry of the Neelum River region, Pakistan.

In Chapter 2, the Great Escarpment of southwestern Saudi Arabia along Red Sea margin is studied in order to elucidate its evolutional/erosional model. Low temperature thermochronologic techniques including (U-Th-[Sm])/He and fission track dating on apatite are used in combination with the geomorphic analysis of the region. This chapter is currently in preparation for submission:

 Turab, S.A., Stüwe, K., Stuart, F.M., Chew, D.M., and Cogné, N., 2016, Escarpment evolution at the Red Sea continental margin of southwestern Saudi Arabia.

In Chapter 3, overall conclusions of this thesis are provided.

In Appendix A, a collection of abstracts from this PhD study is provided that are published/submitted at different international conferences.

 Turab, S.A., Stüwe, K., Stuart, F.M., and Chew, D.M., 2016, An integrated approach to study the exhumation of rocks in Neelum valley, NW Himalayas, Pakistan: Geophysical Research Abstracts, v. 18, EGU2016-4788, EGU General Assembly, 17- 22 April, Vienna, Austria.  Turab, S.A., Stüwe, K., Stuart, F.M., Chew, D.M., and Cogné, N., 2016, Constraining Uplift and Erosion in the Western Himalayan Syntaxis. A Multiple Thermochronometric Study from the Neelum River Region, NW Himalayas, Pakistan. 15th International Conference on Thermochronology (Thermo-2016), 18-23 September, Maresias, Brazil.  Turab, S.A., Stüwe, K., Stuart, F.M., Chew, D.M., and Cogné, N., 2017, Escarpment evolution at the Red Sea continental margin of southwestern Saudi Arabia. Submitted for EGU General Assembly, 23-28 April, Vienna, Austria.

In Appendix B, standard lab procedures, some relevant figures and details about all the samples collected as part of this PhD study are provided.

v

Turab, S.A., 2016

Abstract

In this thesis, I present the results of a study carried out to constrain the amount and understand the processes of rock exhumation at northwest Himalayas and Great Escarpment of southwest Saudi Arabia along the Red Sea margin. A combination of several low temperature thermochronologic techniques such as U-Pb, fission track and (U-Th-[Sm])/He on apatite (AFT and AHe) and (U-Th)/He on zircon (ZHe) along with the geomorphic analysis was used to quantify the rock exhumation. In northwest Himalayas, I selected the Neelum river valley, which is a region missing thermochronological data, but is critical to understand the spatial distribution of exhumation around the Nanga Parbat syntaxis. A total of 13 samples were dated using different combinations of the aforementioned thermochronologic methods. Indeed, the obtained apatite U-Pb data helps constrain the position of the MCT that has remained a topic of controversy. MCT location was also confirmed in the field with the presence of a mylonite zone mapped by earlier researchers. Average ZHe, AFT and average AHe ages are generally 9 – 16 Ma, 3 - 7 Ma and 3 – 6 Ma, respectively, thus indicating exhumation rates of 1 mm/year to 0.3 mm/year. Thermal history modelling confirms the MBT thrusting around 10 Ma and also the MCT inactivity since at-least 11-10 Ma as both the hanging wall and footwall samples show no difference in post 11-10 Ma thermal histories when modelled together as well as separate blocks. Stream power analysis show strong geomorphic disequilibrium caused by recent activity along the MBT/MCT and Knick points identified on long profiles indicate a wave of erosion propagating upstream that may have started as a result of base-level fall at 5-3 Ma. In combination with published AFT ages, our data allow well defined contouring of exhumation ages around the entire western syntaxis region. The age contours trend parallel to the regional-scale thrust faults indicating that exhumation in northwest Himalayas is likely to be controlled by the crustal-scale thrusts rather than fluvial erosion.

In southwest Saudi Arabia, a region was chosen where the escarpment is best developed – a 500 km long section of the escarpment around Saudi Arabia’s highest peak. A total of 22 samples were dated using AFT and AHe methods. AFT and single grain AHe ages range from 13.2 to 352.1 Ma and 2.8 to 264.5 Ma respectively. These ages indicate that there has not been enough (uniform) uplift and erosion throughout the study area to reset all samples. This interpretation is also supported by stream power analysis of the region. However, the reset AFT ages near the coastal plains indicate about 4.5 km of exhumation which although may

vi

Turab, S.A., 2016 have started earlier (early Miocene) but most of it took place after 13.2 Ma. Base of the paleo- PAZ, present at the onset of Red Sea rifting and subsequent uplifting, is interpreted to be at ~200 m present day elevation. Amount of exhumation increases from south towards north and is controlled by erosion in latter and probably by tectonics in former. The distribution of AFT and AHe ages shows that the escarpment evolved by the downwearing of an elevated plateau. Geological evidences suggest that Red Sea rifting did not follow the suggested sequences of events for either active or passive rifting models. Rather, it may have resulted from the complex interaction of these models. The data also help to support an asymmetric rifting model for the Red Sea rift.

In summary, the exhumation of rocks in the two chosen examples for extensional and compressional mountain belts is controlled by dissimilar processes. Tectonics plays the major role in exhuming rocks in northwest Himalayas whereas at the Red Sea escarpment of southwest Saudi Arabia, erosion seems to be a dominant factor, thus implying that results of this study might have global implications to other analogous mountain belts.

vii

Turab, S.A., 2016

Zusammenfassung

In dieser Arbeit präsentiere ich die Ergebnisse einer Studie zum Verständnis der Exhumierung von Gesteinen im nordwestlichen Himalaya und dem Great Escarpment im südwestlichen Saudi-Arabien. Um die Exhumierung zu quantifizieren, wurde eine Kombination mehrerer niedrig-Temperatur thermochronologischer Techniken, wie U-Pb, Fission Track und (U-Th- [Sm])/He in Apatite (AFT und AHe) und (U-Th)/He in Zirkon (ZHe), zusammen mit geomorphologischer Analyse, verwendet. Im nordwestlichen Himalaya wurde das Neelum Flusstal gewählt, eine Region in der thermochronologische Daten fehlen, die allerdings für das Verständnis der räumlichen Verteilung der Exhumierung um die Nanga Parbat Syntaxis entscheidend ist. Insgesamt wurden 13 Proben, mit verschiedenen Kombinationen der oben genannten thermochronologischen Methoden, datiert. Die gewonnenen U-Pb Daten werden benutzt um die Position der Main Central Thrust (MCT) einzuschränken, ein Thema welches noch immer zur Diskussion steht. Im Gelände wurde die MCT Lokalität zusätzlich durch eine von früheren Forschern kartierten, Mylonit-Zone bestätigt. Die durchschnittlichen ZHe, AFT und mittleren AHe Alter sind allgemein 9 – 16 Ma, 3 – 7 Ma und 3 – 6 Ma, und weisen somit auf Exhumierungsraten von 1 mm/Jahr bis 0.3 mm/Jahr hin. Modellierung der thermischen Geschichte bestätigt die Main Boundary Thrust (MBT) Überschiebung um 10 Ma und ebenfalls die Inaktivität entlang der MCT seit zumindest 11 – 10 Ma, da sowohl die Proben der hängenden, als auch der liegenden Blöcke, keinen Unterschied in der thermalen Geschichte nach 11 – 10 Ma Jahren zeigen, unabhängig davon ob die Blöcke zusammen oder getrennt modelliert werden. Stream Power Analyse zeigt ein starkes geomorphisches Ungleichgewicht, welches durch die kürzliche Aktivität entlang der MBT/MCT verursacht wurde, und Knickpunkte, welche entlang der Flussprofile identifiziert wurden, weisen auf eine, flussaufwärts propagierende Erosion hin, welche das Resultat eines Base-level Abfalls vor 5 – 3 Ma sein könnte. In Kombination mit publizierten AFT Altern, erlauben unsere Daten eine gut definierte Konturierung der Exhumationsalter um die gesamte Region der westlichen Syntaxis. Die Alterskontierungen verlaufen parallel zu den, im regionalen Maßstab, Überschiebungen, was darauf hinweist, das Exhumierung in den nordwestlichen Himalayas wahrscheinlicher durch Überschiebungen im Krusten Maßstab als fluviale Erosion kontrolliert ist.

Im südwestlichen Saudi-Arabiens wurde eine Region ausgewählt, in der das Escarpment am besten entwickelt ist - ein 500 km langer Abschnitt um Saudi-Arabiens höchsten Gipfel. Insgesamt wurden 22 Proben mit AFT- und AHe-Methoden datiert. AFT und Einzelkorn AHe viii

Turab, S.A., 2016

Altersgruppen variieren von 13.2 bis 352.1 Ma beziehungsweise 2.8 bis 264.5 Ma. Diese Alter zeigen, dass es nicht genug (einheitlichen) Aufstieg und Erosion im gesamten Studienbereich gab, um die Alter aller Proben zurückzusetzen. Diese Interpretation wird auch mit den geomorphologische Analysen der Region unterstützt. Jedoch zeigen die zurückgesetzten AFT-Alter in der Nähe der Küstenebenen ungefähr 4.5 Kilometer der Exhumierung an, die zwar früher begonnen haben könnten (frühes Miozän), aber das meiste fand nach 13.2 Ma statt. Die Basis des Paläo-PAZ, die am Beginn des Roten Meeres-Rifts und des anschließenden Anhebens vorhanden ist, wird auf ~ 200 m heutiger Höhe ausgelegt. Die Exhumierung nimmt von Süden nach Norden zu und wird durch Erosion in Norden und vermutlich durch Tektonik in Süden kontrolliert. Die Verteilung der AFT und AHe Alter zeigt, dass sich die Steilküste durch die Senkung eines erhöhten Plateaus entwickelte. Geologische Beweise deuten darauf hin, dass das Rote Meer Rifting nicht den vorgeschlagenen Sequenzen von Ereignissen für entweder aktive oder passive Rifting Modelle folgte. Eher kann es aus der komplexen Wechselwirkung dieser Modelle resultieren. Die Daten helfen auch, ein asymmetrisches Rifting Modell für das Rote Meer zu unterstützen.

Zusammenfassend kann man sagen, dass die Exhumierung von Gesteinen in den beiden ausgewählten Beispielen für Dehnungs- und Kompressionsgebirge durch ungleiche Prozesse kontrolliert wird. Die Tektonik spielt die wichtigste Rolle bei der Exhumierung von Gesteinen im nordwestlichen Himalaja, wogegen die Erosion an der Steilküste des Roten Meers im südwestlichen Saudi-Arabien ein dominanter Faktor ist. Somit könnten die Ergebnisse dieser Forschung globale Auswirkungen auf andere analoge Gebirge haben.

ix

Turab, S.A., 2016

Chapter 1.

New constraints on exhumation of the western Himalayan syntaxis. A low temperature thermochronometry of the Neelum River region, Pakistan

1

Turab, S.A., 2016

ABSTRACT

Understanding of the geometry of exhumation in the western Himalayan syntaxis region around Nanga Parbat is of critical importance to the tectonic interpretation of the region. In order to improve our knowledge of this geometry and ultimately assess the role of tectonics in the exhumation of the region we combine apatite and zircon (U-Th-[Sm])/He, apatite fission track and apatite U-Pb dating with geomorphic analysis from a region that is critical to this understanding, but has so far escaped geochronological studies: the Neelum valley region of Azad Jammu and Kashmir, Pakistan. Pooled fission track ages range from 2.2 ± 0.4 to 7.0 ± 0.4 Ma (1σ), apatite He ages range from 2.0 ± 0.2 to 8.7 ± 0.2 Ma and zircon He ages range from 6.1 ± 0.1 to 20.0 ± 0.4 Ma. Exhumation contours run parallel to the Main Boundary- and Main Central-Thrusts. Thermal history modeling indicates that the region experienced accelerated exhumation at 8-10 Ma that removed 5-6 km of crust. This is synchronous with the initiation of movement along Main Boundary Thrust. Apatite U-Pb ages range from Proterozoic to mid-Miocene and can be separated into three groups depending on the degree they have been affected by Himalayan tectonothermal events. Stream power analysis of the 0.9 Neelum River catchment indicates high steepness index of ksn >500 m along the major river and lower ksn in the headwaters. Furthermore, the boundary between the U-Pb age groups, and the transition from high ksn to lower ksn values along the main Neelum River, coincide with the mapped trace of the Main Central Thrust, corroborating the presence of the thrust in the southeastern part of the region. Collectively this data provides convincing support for the contention that tectonics is the main driving mechanism for exhumation at the western syntaxis of the Himalayas.

2

Turab, S.A., 2016

INTRODUCTION

The western syntaxis of the India-Asia Collision zone is arguably the most active and spectacular example of a collisional orogeny. It features extreme examples of both mountain building and mountain erosion. Nanga Parbat, the ninth highest mountain on Earth, lies at its center and one of the largest drainage systems in Asia, the Indus river, also crosses the syntaxis region (Fig. 1.1A). The syntaxis is surrounded by several crustal-scale thrusts (Main Mantle Thrust - MMT, Main Central Thrust - MCT and Main Boundary Thrust - MBT) (Burg et al., 1996; Guillot et al., 2008; Wilke et al., 2012; Pêcher et al., 2002, 2008; Zeitler et al., 2001a,b; Zeitler, 1985; Fig. 1.1B) and it features some of the world’s highest rock uplift and erosion rates (Zeitler, 1985; Zeitler et al., 1982; Wilke et al., 2012). In the Nanga Parbat region, fluvial incision rates of up to 12 mm/year have been measured (Burbank et al., 1996) and approximately 25 km of erosion has taken place since the initiation of collision between India and Asia in the Paleocene (Butler et al., 2002).

Originally, two main drivers for the high exhumation rates have been debated. One hypothesis considers that the rock uplift is directly related to the extreme plate convergence rates (>10 - 15 mm/year) (Khan et al., 2008; Seeber and Pêcher, 1998; Garcia-Castellanos and Jiménez- Munt, 2015). The alternative hypothesis proposes that erosion by rivers provides extreme denudational unloading so that the uplift rate is driven by fluvial erosion. This model implies that there is a direct feedback between deep-seated tectonic and surface processes (Koons et al., 2013; Zeitler et al., 2001a,b). More recent studies have shown that the extreme topography of the region can only be sustained if there is more mass influx into the syntaxis region than due to the normal convergence between India and Asia (Whipp et al., 2014). This recognition has led to the formulation of the “tectonic aneurysm” model (e.g. Koons et al., 2013), a model which implies localized uplift and exhumation in the syntaxis region, geometrically unrelated to either the major drainages or the major thrusts of the region. In short, the cause and consequence of mountain building and erosion is an enduring debate and a good knowledge of the spatial geometry of exhumation in the syntaxis region helps to discern between different models.

3

Turab, S.A., 2016

Figure 1.1: Topographic and geological maps showing the major geologic subdivision of the northwest Himalayas in the western syntaxis region (after Wilke et al., 2012). (A) Topographic map of the western syntaxis with location as inset. The red rectangle in the inset figure shows the location of Fig. 1.1 in regional context. (B) Geological map of the same region as (A). Mint Green = Karakoram (Asian Plate); Pink = Kohistan Ladakh Arc; Turquoise = Higher Himalayas (Crystalline and Intrusive – HHC); Brown = Lesser Himalayas (Metasediments – LH); Yellow = Sub Himalayas (Molasse Sediments). Asterisk shows the epicenter location of the 2005 Kashmir earthquake. (C) Map showing published apatite fission track ages and contours in the region of the Nanga Parbat syntaxis. Green dots and contours are after Zeitler (1985) also showing region of high exhumation rates (green area). Dashed parts of green contours are as shown by Zeitler (1985) but not necessarily supported by data and reinterpreted herein. Red dots are after Wilke et al. (2012) and references therein, Purple dots are from van der Beek et al. (2009) and Black dots in the core of Nanga Parbat-Haramosh massif is general representation of data along Indus Gorge and Astore valley transacts from Treloar et al. (2000). Contours are labelled for apatite fission track ages in million years for base-level samples. Dashed contours (both red and green) show different possibilities how Zeitler’s (1985) contours might extend. MKT - Main Karakoram Thrust; MMT – Main Mantle Thrust; STDZ – South Tibet Detachment Zone; MCT – Main Central Thrust; MBT – Main Boundary Thrust; MFT – Main Frontal Thrust (Salt Range Thrust); H - Haramosh. 4

Turab, S.A., 2016

Despite significant earlier research (e.g. Treloar et al., 1989, 1991, 2000; Zeitler et al., 1982, 2001a,b; Zeitler 1985; Schneider et al., 2001), the regional distribution of the rapid exhumation remains poorly constrained (Fig. 1.1C), in part due to the inaccessibility of some critical regions. Zeitler (1985) contoured the region of high uplift and erosion rates using apatite fission track (AFT) ages (green dots and contours on Fig. 1.1C), and concluded that uplift varied both spatially and temporally with western parts of the syntaxis being uplifted less than 6 km in the last 35 Myr, while the Nanga Parbat and Hunza regions experienced more than 10 km of exhumation in the last 10 Myr (green-shaded region in Fig. 1.1C). However, the existing rock cooling age data (red and purple dots and red dashed contour lines on Fig. 1.1C) and the cooling age summaries of Schneider et al. (2001) or the summary of Koons et al. (2013) allow only tight constraints on the geometry of exhumation near Nanga Parbat itself.

Interpreting the spatial distribution of exhumation rates on the scale of the syntaxis as a whole is hampered by the absence of data from a critical region from which geochronological data are notably absent. This is the region directly south of the Nanga Parbat syntaxis around the Neelum valley region of Azad Jammu and Kashmir, Pakistan (Figs. 1.1C, 1.2). In order to quantify the amount and rates of Late Cenozoic exhumation in this region we have performed a detailed study of the low temperature thermochronology (e.g. Kirstein et al., 2006) in the Neelum valley region. We combine apatite and zircon (U-Th-[Sm])/He, and apatite fission track and U-Pb thermochronometry from two vertical profiles and an along-river transect in the Neelum river catchment and complement this dataset with geomorphic analyses of selected fluvial channels from the region. This data is then interpreted in terms of the geomorphological and geodynamic evolution of the Nanga Parbat syntaxis as a whole.

GEOLOGICAL BACKGROUND

The Himalayan mountain belt formed due to the complex collision of the Indian and Asian plates that started at 65 - 55 Ma (Kazmi and Jan, 1997). In the northwest Himalayas the main rock units are separated by regional-scale thrust faults (Fig. 1.1B). In the northern-most parts of the region, the Main Karakoram Thrust (MKT) separates rocks of the Karakoram micro- plate in its hanging wall from the rocks belonging to Kohistan-Ladakh island arc in the footwall. The Kohistan-Ladakh island arc is principally composed of granodiorite to granite plutons of middle Cretaceous (110 – 90 Ma) age with some metasedimentary rocks (Rolland,

5

Turab, S.A., 2016

2002). The Main Mantle Thrust (MMT) demarcates the southern boundary of the Kohistan- Ladakh arc with the High Himalayan Crystalline (HHC) rocks in the footwall. HHC rocks are mainly high-grade metamorphic rocks intruded by Mansehra granite (Calkins et al., 1975; Fontan et al., 2000) that have yielded a Rb-Sr whole rock age of 516 ± 16 Ma (Le Fort et al., 1980). The HHC rocks were thrust onto the Lesser Himalayan rocks along the Main Central Thrust (MCT) during the early Miocene (Hodges et al., 1992). The Lesser Himalayan units are composed primarily of low-grade metamorphic and sedimentary rocks. Further south, the Main Boundary Thrust (MBT) has transported rocks of the Lesser Himalayas onto the Sub- Himalayan molasse sediments that were deposited in the Neogene after the initiation of the main stage of the Himalayan orogeny (Kazmi and Jan, 1997; Shah, 2009). Finally, the Main Frontal Thrust (MFT) has transported the entire sequence of rocks over the Quaternary sediments of the Punjab Plain.

Deformation in the Himalayas has generally propagated from north to south. Asia and the Kohistan island arc were sutured in northern Pakistan during late Cretaceous times (102 - 85 Ma) along the MKT (Treloar et al., 1989). Collision of India with Asia and the Kohistan- Ladakh island arc was not synchronous: India collided in northern Pakistan with the Kohistan- Ladakh island arc along the MMT at 65-55 Ma (Kazmi and Jan, 1997), but only at 54 Ma in the Ladakh region of northwest India (Patriat and Achache, 1984). During Oligocene- Miocene, High Himalayan rocks were exhumed from mid-crustal levels along the MCT between 24 Ma and 10 Ma (Searle et al., 2008). Deformation then shifted to lower structural levels to the south. Movement on the MBT initiated in the late Miocene (>10 Ma) in the western Himalaya (Meigs et al., 1995), followed by the development of the Main Frontal Thrust (or Salt Range Thrust), the youngest of all the regional faults, at 5.4-1.0 Ma (Kazmi and Jan, 1997). The study region, the Neelum valley, lies largely between the MCT and the MMT and is comprised of the High Himalayan Crystalline (HHC) units.

Despite this well-defined sequence of major tectonic activity along the major thrusts, it is not clear if the final exhumation of the upper crust rocks and the topography of the present day landscape are related to the same series of thrusting events. For example, although the main movement along the MCT is thought to have taken place during the early to middle Miocene (Butler et al., 1989; Vannay et al., 2004), apatite fission track ages from HHC rocks in the syntaxis region are substantially younger (Zeitler et al., 1982; Zeitler, 1985). Generally these ages are ~1 Ma in the core region of Nanga Parbat syntaxis, whereas they are as old as 6 Ma towards the southwest in the Kaghan and Indus valleys (Fig. 1.1C). Wilke et al. (2012) report

6

Turab, S.A., 2016

AFT ages of ≤10 Ma from the upper Kaghan valley, but their samples are not from the lowest points in the landscape so they do not record the latest thermal history. Although the MMT may only have remained active from early Eocene until late Oligocene or early Miocene (DiPietro and Pogue, 2004), many of the low-temperature cooling ages from the Kohistan arc are significantly younger (10–20 Ma, e.g. Zeitler, 1985). Thus, much of the final stages of exhumation, and probably also the surface uplift, is younger than the compressional tectonic history that has been determined by geochronology, and paleomagnetic, metamorphic and stratigraphic methods (Treloar et al., 1989; Kazmi and Jan, 1997; Patriat and Achache, 1984; Searle et al., 2008; Meigs et al., 1995).

The geology of the Neelum valley region was mapped by Malik et al. (1996) and Fontan et al. (2000). The region lies in the hanging wall of MBT and forms three distinct tectonic units separated by two thrusts: (i) the Gumot shear zone (GSZ) which is a 100-200 m wide ductile shear zone inside the HHC, and (ii) the MCT only some kilometers in the hanging wall of MBT. From the structurally highest to lowest levels the tectonic units separated by these thrusts are the Shontar unit in the hanging wall of GSZ, the Kalapani unit between the GSZ and MCT, and the Salkhala unit below the MCT (inset in Fig. 1.2). The GSZ is only a local structure within the HHC in the Neelum valley region. Although the relationship between the MCT and the syntaxis as a whole is clear, the position of the MCT in the Neelum valley region as shown on Fig. 1.2 is subject to debate. Fontan et al. (2000) proposed that the MCT crosses the Neelum River near Doarian village (Fig. 1.2). In contrast, Greco and Spencer (1993) mapped a mylonite zone along MCT that runs through the Leswa section, crossing the Neelum River between Nauseri and Jura (Fig. 1.2). This widely acknowledged position (Wilke et al., 2012; Pêcher et al., 2008; Searle et al., 2008; Guillot et al., 2008; DiPietro and Pogue, 2004; Greco and Spencer, 1993) of MCT is shown in Fig. 1.1. During fieldwork for this study we observed a mylonite zone at a similar location along the Leswa road. The timing of thrusting on these structures is not known and is correlated with movement on the MCT in other portions of the northwest Himalayas where it was active until 16 Ma (Vannay et al., 2004) and possibly reactivated later (Catlos et al., 2001; Hodges et al., 2004). The morphology of the Neelum river catchment is similar to the Indus valley in the northwest. It spans an elevation range from 700 m (at the junction with the Jhelum river) to about 5,000 m, and crosses almost the entire morphological range of the Himalayas from the Punjab plains to the high Deosai plateau which is thought to be the preserved remnant of the Eocene Tibetan Plateau in the east (Van der Beek et al., 2009).

7

Turab, S.A., 2016

Figure 1.2: Geological map of the Neelum valley region with locations of samples collected for this study (red and green dots) (after Fontan et al., 2000). Inset shows the broader structural subdivisions of the area. Note that (still controversial) MCT is marked at different location in this map by Fontan et al. (2000) as compared to others (e.g. Greco and Spencer, 1993; DiPietro and Pogue, 2004; Wilke et al., 2012; Pêcher et al., 2008; Searle et al., 2008; and Guillot et al., 2008, and as shown on Fig. 1.1).

LOW TEMPERATURE THERMOCHRONOLOGY: SAMPLES AND TECHNIQUES

Rock samples were collected along high-relief sections from near the Neelum River to the adjoining peaks in the Neelum valley. The samples were collected near major regional-scale structures (e.g. MCT and MBT) along two high relief sections (Fig. 1.2); one near Kel in the upper parts of the Neelum River valley and the second along the Leswa road in the lower part. The Leswa road leaves the main Muzaffarabad-Kel road in the Neelum River valley, crosses a pass and joins the valley again near Maira (Fig. 1.2). Special attention was also paid to the selection of sampling sites that cover maximum altitudinal range with minimum horizontal

8

Turab, S.A., 2016 separation between samples of the same profile. Samples were also collected between the two transects along the Neelum River to document changes in exhumation rate.

Ten samples were collected between 2,000 m and 4,200 m elevation along the Kel relief section. The average vertical separation between samples is 200-250 m (Table 1.1). Seven samples were collected along Neelum River between the Kel and Leswa sections at ~100 m vertical separation. Thirteen samples were collected along the Leswa relief section. Samples were taken from both sides of the Leswa-pass between ~1,200 m at the Neelum River and ~2,700 m at the pass (Table 1.1). Two samples are sandstones of the Murree formation, one is a sample of Panjal andesite and the remaining ten samples are metamorphic rocks. The quality and quantity of apatite was variable so only a sub-set of the samples could be used for analysis. Where possible we have combined apatite and zircon (U-Th-[Sm])/He, apatite fission track and apatite U-Pb in order to obtain the regional cooling history from ~450 °C to ~40 °C.

TABLE 1.1: DETAILS OF SAMPLES TAKEN FOR LOW TEMPERATURE THERMOCHRONOLOGIC ANALYSIS Sample Latitude* Longitude* Sample reference Elevation Rock type (°N) (°E) (m) Garnet bearing Gneiss, Medium K1 34 51 30.5 74 17 52.4 Kel 4225 grained K2 34 51 14.5 74 17 46.3 Kel 3916 Garnet bearing Gneiss, Pegmatite K3 34 51 01.9 74 17 47.1 Kel 3768 Pegmatite K6 34 50 23.7 74 18 03.7 Kel 3040 Pegmatite, Garnet bearing K10 34 49 12.9 74 18 22.5 Kel / Neelum 2015 Mica Gneiss, Garnet present N5 34 38 52.1 73 56 13.7 Neelum 1469 Granite N7 34 32 04.9 73 50 45.1 Neelum 1285 Granite N8 34 27 46.1 73 49 11.7 Neelum / Leswa 1177 Pegmatitic Gneiss L2 34 27 55.2 73 48 36.9 Leswa 1357 Granitic Gneiss L6 34 27 41.2 73 45 00.7 Leswa 2329 Gneiss (Mica-Bio + Mus) L7 34 27 25.8 73 44 51.1 Leswa 2544 Gneiss L9 34 26 18.2 73 45 04.8 Leswa 2489 Gneiss L10 34 25 43.7 73 44 15.6 Leswa 2293 Gneiss *data entered in “Degrees Minutes Seconds” format.

Sixteen samples were selected for analysis. Apatite from 12 samples were dated using apatite fission track (AFT) and U-Pb methods (Table 1.2), which comprised four samples from the Kel profile line, three samples from the Neelum river and five from the Leswa profile line. Eight samples (largely a subset of the 12 used for AFT and U-Pb dating) were dated by the 9

Turab, S.A., 2016 apatite U-Th-[Sm]/He (AHe) method (Farley, 2002) (Table 1.3) and seven samples were dated by the zircon U-Th/He (ZHe) method (Reiners, 2005) (Table 1.4).

The thermal history of the three sample groups (Kel, Leswa and Neelum river profiles) have been modeled using the software QTQt version 5.3.3 (Gallagher, 2012). Information from all thermochronometers was used for modelling the thermal histories. This incorporated the fission track annealing model of Ketcham et al. (2007) and the effect of radiation damage on He ages using the models of Flowers et al. (2009) for apatite and Guenthner et al. (2013) for zircon. We used the weighted mean U-Pb age of all the samples that have Cenozoic lower intercept ages (unanchored) as a high-temperature constraint: 28 ± 6 Ma and 450 ± 75 °C. The maximum temperature change (heating or cooling rate) for all model runs was limited to 1000 °C/Myr. The present-day temperature offset between the top and bottom samples of the Kel and Leswa profiles was set at 15 ± 5 °C and 10 ± 5 °C and a present-day temperature of 5 ± 5 °C and 10 ± 10 °C was used for the top sample in each profile, respectively. The palaeo- temperature offset is based on the total vertical separation between the top and bottom samples of a profile section. The models were run for 200,000 iterations as burn-in (which were discarded) and the subsequent 200,000 iterations were used for the analysis.

THERMOCHRONOLOGICAL RESULTS

AFT ages range from 2.2 ± 0.4 to 7.0 ± 0.4 Ma (1σ) (Fig. 1.3, Table 1.2). Only sample K2 had sufficient confined tracks (n = 44) to allow track length measurements. It has a mean track length (MTL) of 13.36 ± 1.65 µm. The kinetic parameter, Dpar, shows a range from 1.11 to 1.59 µm, while the chlorine content ranges from 0.01 to 0.74 wt %. All samples passed the chi-squared (χ2) test (Galbraith, 1981, modified for LA-ICP-MS fission track dating by Donelick et al., 2005).

AFT ages at the base (2,015 m) and top (3,916 m) of the Kel profile are ~4 Ma and ~7 Ma respectively and show a positive correlation with elevation thus giving an average exhumation rate of 0.6 mm/yr through apatite partial annealing zone (PAZ) for these samples (Fig. 1.3). AFT ages of the Leswa profile show no resolvable relationship with altitude, ranging from 2.2 to 3.5 Ma. The ages of the six samples are indistinguishable within analytical uncertainty indicating rapid exhumation. The majority of the samples from the Neelum River yield ages between 3.6 and 4.0 Ma and are indistinguishable within analytical uncertainty. Sample N8, from the base of the Leswa profile gives a slightly younger cooling age (Fig. 1.3). 10

Turab, S.A., 2016

Figure 1.3: Apatite fission track thermochronology done for this study shown as Plot of AFT ages versus elevation. Note that sample K10 is shown in blue (i.e. part of the Kel section), but is at Neelum river level. Its value can therefore also be read in connection with the ages shown in red (Neelum river samples). Corresponding logic applies to sample N8 (see sample locations on Fig. 1.2).

AHe ages range from 1.4 ± 0.13 to 7.6 ± 0.2 Ma (1σ, uncorrected) and from 2.0 ± 0.2 to 8.7 ± 0.2 Ma (1σ, FT-corrected) (Fig. 1.4; Table 1.3). Average apatite He (Ave. AHe) ages typically overlap or are slightly younger than the corresponding AFT ages (Fig. 1.4). As with the AFT data, there is a weak trend of increasing AHe ages with elevation for the Kel profile, but no significant AHe age deviation with elevation for the Leswa profile is observed (Fig. 1.4; Table 1.3). The old AHe ages from sample L6 are not used in the calculation of an average age as these were generally poor quality apatite grains.

Individual uncorrected ZHe ages range from 5.2 ± 0.1 to 13.3 ± 0.3 Ma (1σ), while corrected ages range from 6.1 ± 0.1 to 20.0 ± 0.4 Ma (1σ) (Fig. 1.4; Table 1.4). Within-sample age variation is significant, typical of many ZHe studies. This may reflect U and/or Th zonation (Dobson et al., 2008) and radiation damage (e.g. Guenthner et al., 2013). Average ZHe ages range from 8.8 ± 0.7 to 16.5 ± 2.2 Ma and there is no significant variation with elevation.

11

Turab, S.A., 2016

TABLE 1.2: APATITE FISSION TRACK AND U-Pb RESULTS 2 2 Sample Number Ns ΣpiΩi 1σ ΣpiΩi χ P(χ ) Cl 1σ Dpar 1σ Pooled FT U-Pb TW intercept age U-Pb TW intercept * of wt% Cl (µm) Dpar age (unanchored) age (anchored) grains wt% (µm) (Ma ± 1σ) (Ma ± 2σ) (Ma ± 2σ) K2 26 574 7.901E-04 1.031E-05 36.18 0.07 0.73 0.22 1.55 0.23 6.9 ± 0.3 23.7 ± 7.2 18.5 ± 4.6 K3 48 380 5.142E-04 5.440E-06 44.43 0.58 0.18 0.23 1.59 0.26 7.0 ± 0.4 17.0 ± 10.0 24.3 ± 8.0 K6 51 52 1.128E-04 1.065E-06 42.88 0.75 0.01 0.25 1.33 0.18 4.4 ± 0.6 32.0 ± 26.0 30.0 ± 18.0 K10 49 127 3.052E-04 9.528E-07 54.33 0.25 0.05 0.09 1.40 0.18 3.9 ± 0.4 19.0 ± 12.0 43.0 ± 12.0 N5 45 192 4.817E-04 4.735E-06 51.20 0.21 0.07 0.15 1.31 0.03 3.8 ± 0.3 25.0 ± 12.0 36.1 ± 8.0 N7 48 109 2.784E-04 2.554E-06 27.90 0.99 0.04 0.05 1.11 0.18 3.7 ± 0.4 42.0 ± 11.0 48.3 ± 9.1 N8 51 77 2.501E-04 3.177E-06 34.53 0.95 0.03 0.05 1.13 0.19 2.9 ± 0.3 20.0 ± 19.0 29.0 ± 16.0 L2 49 230 6.230E-04 1.062E-05 35.27 0.91 0.06 0.07 1.21 0.22 3.5 ± 0.2 30.0 ± 4.4 28.9 ± 4.0 Mainly Palaeoproterozoic L6 43 40 1.271E-04 1.138E-06 38.67 0.62 0.01 0.07 1.13 0.17 3.0 ± 0.5 N.D.† apatite; no young grains Mainly Palaeoproterozoic L7 50 25 1.082E-04 1.055E-06 46.19 0.59 0.09 0.14 1.28 0.03 2.2 ± 0.4 N.D.† apatite; no young grains Mainly Palaeoproterozoic L9 46 65 1.853E-04 1.721E-06 42.02 0.60 0.05 0.12 1.34 0.16 3.3 ± 0.4 N.D.† apatite; no young grains Predominantly c. 750 Ma L10 49 42 1.584E-04 1.896E-06 31.23 0.97 0.10 0.14 1.39 0.03 2.5 ± 0.4 N.D.† apatite Note: Details about methodology and symbols are given in Supplementary material and references therein. Zeta calibration value (ζ) is 18.92 ± 0.36. *Anchored at a 207Pb/206Pb initial value of 0.837 ± 0.005 calculated from Stacey and Kramers (1975) crustal Pb evolution model (see text for details). †N.D. = not determined.

12

Turab, S.A., 2016

Figure 1.4: Plots for Kel, Neelum river and Leswa samples showing relationship between ZHe, AFT and AHe ages, plotted against the sample elevations. Note that the Neelum river plot shares samples K10 and N8 with Kel and Leswa plots, respectively (see text for details). Samples not used for calculation of average ages are shown in the lighter shade of the same color. Average sample AHe and ZHe ages are shown by a larger symbol of the same color. 13

Turab, S.A., 2016

TABLE 1.3: APATITE (U-Th-Sm)/He RESULTS Sample Mass U Th Sm 4He eU Raw Age FT Corrected Age Average Age Th/U Error Diameter (µg) (ppm) (ppm) (ppm) (nmol/g) (ppm) (Ma ± 1σ) (Ma ± 1σ) (Ma ± 1σ) (µm) K1a 0.87 5.0 0.0 241.2 0.1 5.0 3.1 ± 0.7 0.63 4.9 ± 1.1 N.A.* 0.00 0.00 61.7 K2a 32.77 130.0 6.5 543.4 3.9 131.6 5.4 ± 0.2 0.87 6.2 ± 0.3 0.05 0.00 191.7

K2b 42.16 117.0 6.3 33.2 3.6 118.5 5.6 ± 0.4 0.88 6.4 ± 0.4 0.05 0.00 210.0

K2c 34.92 115.7 9.7 499.7 3.4 118.0 5.3 ± 0.3 0.88 6.0 ± 0.4 0.08 0.01 230.0

K2d 63.78 296.2 42.0 404.3 6.5 306.0 3.9 ± 0.3 0.90 4.4 ± 0.3 0.14 0.01 258.3

K2e 2.55 132.8 13.5 15.3 3.3 136.0 4.5 ± 0.3 0.72 6.2 ± 0.4 5.8 ± 0.8 0.10 0.01 96.7 K3a 1.06 47.8 1.5 209.3 0.6 48.2 2.3 ± 0.2 0.64 3.7 ± 0.2 0.03 0.06 60.0

K3b 2.21 32.8 37.5 100.6 1.0 41.6 4.6 ± 0.2 0.70 6.7 ± 0.2 1.15 0.07 83.3

K3c 24.97 40.2 3.1 186.4 1.5 40.9 6.9 ± 0.2 0.87 8.0 ± 0.2 0.08 0.00 223.3

K3d 28.41 44.2 0.3 280.9 0.9 44.3 3.9 ± 0.1 0.87 4.4 ± 0.1 5.7 ± 2.0 0.01 0.00 215.0 K6a 4.51 4.5 2.9 190.7 0.0 5.2 1.5 ± 0.1 0.76 2.0 ± 0.2 0.65 0.14 115.0

K6b 5.46 8.4 5.4 185.0 0.1 9.6 2.3 ± 0.2 0.78 3.0 ± 0.2 0.65 0.09 133.3

K6c 3.06 8.0 8.4 99.3 0.1 10.0 1.6 ± 0.1 0.72 2.3 ± 0.2 1.06 0.13 91.7

K6d 12.44 7.8 1.6 114.6 0.2 8.2 5.5 ± 0.3 0.82 6.7 ± 0.4 0.21 0.03 145.0

K6e 2.08 13.8 3.4 227.5 0.1 14.6 1.4 ± 0.1 0.69 2.0 ± 0.2 3.2 ± 2.0 0.25 0.10 76.7 K10a 5.77 16.2 2.3 91.7 0.4 16.7 4.0 ± 0.2 0.79 5.1 ± 0.3 0.14 0.03 105.0

K10b 4 9.4 1.4 67.7 0.1 9.8 2.3 ± 0.2 0.75 3.1 ± 0.2 0.15 0.07 101.7

K10d 3.08 20.5 2.4 143.6 0.4 21.1 3.5 ± 0.2 0.74 4.7 ± 0.2 4.3 ± 1.1 0.12 0.04 86.7 N8b 3.8 1.3 4.4 23.4 0.0 2.3 2.6 ± 0.4 0.74 3.5 ± 0.5 3.38 0.80 113.3

N8c 3.9 4.3 31.7 47.0 0.1 11.8 1.7 ± 0.1 0.74 2.3 ± 0.1 2.9 ± 0.9 7.40 0.43 95.0 L6a 98.19 5.2 11.7 293.2 0.2 8.0 5.0 ± 0.1 0.91 5.5 ± 0.1 2.26 0.06 296.7

L6b 14.59 18.3 34.0 432.0 0.3 26.3 2.4 ± 0.1 0.84 2.8 ± 0.1 1.87 0.13 163.3

L6c 31.98 12.2 26.1 484.8 0.8 18.3 7.6 ± 0.2 0.88 8.7 ± 0.2 2.16 0.07 233.3

L6d 67.21 4.8 10.5 321.6 0.2 7.2 5.4 ± 0.1 0.90 6.0 ± 0.1 2.23 0.05 271.7

L6e 43.38 6.1 23.6 296.4 0.3 11.6 5.0 ± 0.1 0.88 5.7 ± 0.1 5.7 ± 2.1 3.91 0.09 231.7 L9a 31.12 17.9 23.2 322.8 0.3 23.4 2.4 ± 0.1 0.88 2.8 ± 0.1 1.30 0.03 206.7

L9c 3.55 49.8 31.9 314.0 0.6 57.4 1.8 ± 0.1 0.73 2.5 ± 0.1 2.6 ± 0.2 0.65 0.05 88.3 Note: Average ages and associated errors are calculated as mean and standard deviation of all the aliquots from the same sample. *N.A. = not applicable

14

Turab, S.A., 2016

TABLE 1.4: ZIRCON (U-Th)/He RESULTS Sample Mass U Th 4He eU Raw Age FT Corrected Age Average Age Th/U Error Diameter (µg) (ppm) (ppm) (nmol/g) (ppm) (Ma ± 1σ) (Ma ± 1σ) (Ma ± 1σ) (µm) K3vi 1.81 1431.3 15.4 69.5 1434.9 9.0 ± 0.2 0.69 13.0 ± 0.3 0.01 0.00 57.5

K3vii 3.97 1489.5 22.2 64.3 1494.7 8.0 ± 0.2 0.78 10.3 ± 0.2 0.01 0.00 71.6

K3viii 3.19 2112.1 267.4 80.5 2175.0 6.9 ± 0.2 0.74 9.2 ± 0.2 0.13 0.00 63.2

K3ix 3.57 2749.8 21.2 187.2 2754.8 12.6 ± 0.3 0.74 17.0 ± 0.4 0.01 0.00 68.8

K3x 15.63 5878.5 77.0 418.4 5896.6 13.2 ± 0.3 0.86 15.2 ± 0.4 12.9 ± 3.2 0.01 0.00 129.2 K6vi 2.67 2010.1 62.1 104.0 2024.7 9.5 ± 0.2 0.67 14.3 ± 0.3 0.03 0.00 50.9

K6vii 1.99 1098.8 47.5 68.3 1110.0 11.4 ± 0.2 0.67 17.0 ± 0.4 0.04 0.00 50.0

K6viii 2.62 2130.0 79.5 129.9 2148.7 11.2 ± 0.2 0.69 16.3 ± 0.3 0.04 0.00 51.8

K6ix 1.95 950.5 62.4 69.2 965.1 13.3 ± 0.3 0.66 20.0 ± 0.4 0.07 0.00 48.6

K6x 1.89 2837.5 120.3 152.1 2865.7 9.8 ± 0.2 0.66 14.9 ± 0.3 16.5 ± 2.2 0.04 0.00 48.2 K10ii 4.91 413.3 127.3 20.2 443.2 8.4 ± 0.2 0.79 10.7 ± 0.2 0.31 0.01 88.2

K10iii 13.29 670.9 581.6 31.1 807.5 7.1 ± 0.1 0.85 8.4 ± 0.1 0.87 0.02 134.2

K10iv 2.19 722.0 101.0 26.9 745.8 6.7 ± 0.1 0.69 9.7 ± 0.2 0.14 0.00 54.0

K10vi 7.29 461.9 122.2 25.7 490.7 9.7 ± 0.2 0.79 12.3 ± 0.2 0.27 0.01 80.9

K10vii 4.13 574.0 143.4 32.5 607.7 9.9 ± 0.2 0.76 13.0 ± 0.3 10.8 ± 1.9 0.25 0.01 74.5 N5i 4.67 546.9 95.5 28.6 569.3 9.3 ± 0.2 0.78 12.0 ± 0.2 0.18 0.00 79.7

N5ii 4.67 725.5 345.4 45.1 806.6 10.3 ± 0.2 0.78 13.3 ± 0.2 0.48 0.01 83.6

N5iii 5.23 854.7 185.0 41.8 898.2 8.6 ± 0.2 0.80 10.8 ± 0.2 0.22 0.01 81.5

N5iv 8.64 773.2 97.2 40.0 796.0 9.3 ± 0.2 0.82 11.4 ± 0.2 0.13 0.00 97.6

N5v 3.05 938.6 122.2 56.5 967.3 10.8 ± 0.2 0.74 14.6 ± 0.3 12.4 ± 1.5 0.13 0.00 67.5 N8i 4.54 2183.1 185.4 79.3 2226.6 6.6 ± 0.1 0.77 8.5 ± 0.2 0.09 0.00 78.8

N8ii 6.39 2619.1 106.1 98.7 2644.1 6.9 ± 0.2 0.80 8.7 ± 0.2 0.04 0.00 86.9

N8iii 1.41 2657.5 107.8 76.3 2682.9 5.3 ± 0.1 0.67 7.9 ± 0.2 0.04 0.00 51.7

N8iv 2.95 918.9 231.6 49.6 973.3 9.4 ± 0.2 0.74 12.8 ± 0.3 9.5 ± 2.2 0.25 0.01 66.7 L6i 66.18 724.8 198.5 40.7 771.5 9.8 ± 0.7 0.90 10.8 ± 0.7 0.28 0.03 191.7

L6ii 23.99 682.0 228.5 33.7 735.7 8.5 ± 0.2 0.86 9.9 ± 0.2 0.34 0.01 126.3

L6iii 20.13 329.7 179.5 17.6 371.9 8.8 ± 0.2 0.86 10.3 ± 0.2 0.55 0.01 131.1

L6iv 19.67 1156.2 648.8 36.8 1308.7 5.2 ± 0.1 0.85 6.1 ± 0.1 0.57 0.01 120.8

L6v 53.33 489.6 256.3 22.2 549.9 7.5 ± 0.1 0.91 8.2 ± 0.2 9.1 ± 1.9 0.53 0.01 196.4 L9ii 9.52 917.5 125.8 36.1 947.1 7.1 ± 0.1 0.83 8.6 ± 0.2 0.14 0.00 107.6

L9iii 18.88 929.5 338.4 44.6 1009.0 8.2 ± 0.2 0.86 9.6 ± 0.2 0.37 0.01 126.5

L9iv 11.44 909.3 112.4 34.6 935.8 6.9 ± 0.2 0.84 8.2 ± 0.2 8.8 ± 0.7 0.12 0.00 118.6 Note: Average ages and associated errors are calculated as mean and standard deviation of all the aliquots from the same sample. 15

Turab, S.A., 2016

Apatite U-Pb data (uncorrected for common Pb) were fitted by isochrones on a Tera– Wasserburg (TW) Concordia to determine lower intercept 238U/206Pb ages (Fig. 1.5). The data fall into three groups: (i) apatites that lie on a well-defined discordia with Cenozoic lower intercepts that range from 17.0 ± 10.0 to 32.0 ± 26.0 Ma (2σ, unanchored) or 18.5 ± 4.6 to 43.0 ± 12.0 Ma (2σ, anchored at a 207Pb/206Pb initial value of 0.837 ± 0.005) (Fig. 1.5A; Table 1.2). (ii) Samples which yield a Cenozoic lower intercept ranging from 20.0 ± 19.0 to 42.0 ± 11.0 Ma (2σ, unanchored) or 28.9 ± 4.0 to 48.3 ± 9.1 Ma (2σ, anchored) but also contain older grains that can be clearly identified as a separate group (Fig. 1.5B). It is important to note that only the younger grains were used for fitting the isochrones. The young apatite age population was also identified independently on the basis of the apatite trace-element chemistry. For example, the apatites yielding the Cenozoic lower intercept in sample N5 (Fig. 1.5B) are depleted in La relative to the older, partly reset grains. (iii) Apatites which yield Proterozoic lower intercepts (Fig. 1.5C).

For all three groups, the first U-Pb age column in Table 1.2 lists “unanchored” lower intercept TW Concordia ages where the isochron regression was constrained by the analyses alone. Lower-intercept TW Concordia ages are then listed in Table 1.2 “anchored” at a 207Pb/206Pb initial value of 0.837 ± 0.005, which represents the common Pb composition at 28 Ma calculated by the Stacey and Kramers (1975) crustal Pb evolution model. The weighted mean age of all the unanchored Cenozoic lower intercept ages is 27.3 ± 5.6 Ma (Fig. 1.5D).

MORPHOLOGICAL ANALYSIS

As a further support to the data presented above we provide some morphological analysis of selected fluvial channels in the region. Channel profile analysis was carried out in ArcMap 10.1 using a 1 arc second (~30 m) resolution Digital Elevation Model (DEM) data (Fig. 1.6). Following common procedure, we study channel characteristics in comparison to the geomorphic steady state where channel slope (S) and drainage area (A) are related by a power -θ law function S= ks A where ks and θ are referred to as steepness and concavity indices, respectively (e.g. Wobus et al., 2006). The values of ks and θ can be measured directly from DEM data by performing a regression analysis on log–log plot between A and S. The normalized steepness index (ksn) is calculated using a fix reference concavity value (θref) which is usually taken around 0.45 (Wobus et al., 2006). The ksn is proportional to erosion rate and can thus be used to infer zones of variable exhumation related to aspects of fault

16

Turab, S.A., 2016

activity (Whittaker, 2012). The ksn of all the streams was calculated in Matlab R2011b by linking it to the ArcMap 10.1 through the “Profiler toolbar” using a value of 0.45 for θref (for details see Whipple et al., 2007).

Figure 1.5: Representative examples of U-Pb Tera-Wasserburg concordia lower-intercept ages for each of the three resultant groups. (A) Example for the completely affected samples from Kel profile in the hanging wall of GSZ; (B) Example of a partially affected sample from within the HHC (red data points are not used for the calculation of the intercept); (C) Example for an unaffected sample from the Leswa profile in the Lesser Himalaya. Discordia diagrams for all other measured samples are shown in Appendix Figure A1. (D) Weighted mean age for all the affected samples. (E) Schematic map of the Neelum river region (after Fontan et al., 2000) showing the distribution of completely, partially and unaffected samples (yellow ellipses; see also Fig. 1.2 for more detail). 17

Turab, S.A., 2016

0.9 The ksn values of almost all the streams are lowest near the headwaters (ksn < 200 m ) and also in the lowest reaches of the Neelum river (yellow parts of streams on Fig. 1.6A). In 0.9 contrast, the central parts of the Neelum river are characterized by high ksn (ksn > 500 m ; red parts of streams on Fig. 1.6A). The transition from high to low ksn values along the Neelum river main channel occurs at the location where it crosses the MBT and MCT (sensu Greco and Spencer, 1993). Whereas Kunhar and Jhelum rivers show comparatively lower ksn values 0.9 (200-500 m ) in their main channels except small segments of higher ksn values primarily close to the regional faults and thus change in lithology (Figs. 1.6A, 1.1B). In summary, the variable steepness index of the Neelum river catchment bears all the characteristics of a catchment that is not in geomorphic equilibrium.

Channel profiles of selected rivers in the region are consistent with the ksn values discussed above. Fig. 1.6B shows the channel profiles for four rivers that are considered to bear critical information for our interpretation below: The Neelum river is of principal interest to our study. The Jhelum river which joins the Neelum from the east just below the MBT is the principal river draining the Kashmir basin to the south of the study region. It was chosen because the history of its confluence will be inferred from the thermochronological data as discussed below. The Kunhar river runs roughly parallel to the Neelum valley to the west and north. It was chosen because its confluence with the Jhelum is only 10 kilometers downstream of the Jhelum-Neelum confluence and because it is expected to bear similar morphological channel characteristics as the latter. Finally, the Sind river that flows from the Zoji La Pass west along the Srinagar – Leh Highway into the Kashmir basin was chosen because it does not cross any of the major faults that dissect the Neelum valley region and can be used for comparison.

Knick points have been identified on the long profiles of Kunhar, Jhelum, Neelum and Sind rivers (Fig. 1.6B). Confluence of Neelum and Kunhar rivers with Jhelum river both occur within about 10 km of each other at the western limb of major antiformal structure of northwestern Himalayas, the Hazara-Kashmir syntaxis (HKS). These three rivers show knick points along their long profiles at a distance of ~90 - ~110 km upstream from the confluence. However, the Sind river, which does not cross any of the major Himalayan structures and joins Jhelum river in Kashmir basin, also show a nick point at a distance of ~80 km upstream where the river long profile shows change from apparently equilibrated channel (downstream) to stepped morphology (upstream). Knick points identified here show a good match with the ksn values of the corresponding rivers (Fig. 1.6).

18

Turab, S.A., 2016

Figure 1.6: Results of geomorphic analyses of the western Himalayan syntaxis area including (A) topographic map and results of a numerical analysis of the stream power, and (B) long profiles of selected streams (i.e. Jhelum, Neelum, Kunhar and Sind rivers). Stream power is color coded for steepness index, normalized to a reference concavity index of 0.45. White stars on map and blue plus signs on long profiles show major knick points. Blue polygon outlines the Neelum river catchment. 19

Turab, S.A., 2016

DISCUSSION

The thermochronometric analyses provide a series of ages that document different parts of the cooling history. Starting with the highest measured closure temperature (U-Pb system in apatite) it was shown that the samples fall into three distinct groups with respect to the possibility to fit them by isochrones. We interpret these three groups in terms of the effects of Himalayan exhumation as (i) completely reset, (ii) partially reset and (iii) unreset ages. One example of each is shown in Fig. 1.5.

All samples from the Kel section in the hanging wall of GSZ show entirely reset apatite U-Pb ages. Hence, assuming a surface temperature of 20 °C and closure temperature of 450 °C for U-Pb system in apatite, a total exhumation of more than 14 km has occurred for the highest-, and more than 16 km for the lowest- sample of the Kel profile. This exhumation occurred between 50 and 10 Ma. According to the map of Fontan et al. (2000) these rocks form part of the HHC suite from the hanging wall of the GSZ. The partially reset samples of the second group all come from the lower part of the Neelum valley and comprise of granites to granitic gneisses (Table 1.1; Figs. 1.2, 1.5E). These analyses may represent old (Proterozoic?) apatites that have been partly to wholly reset during Himalayan orogenesis, with possible additional growth of new metamorphic apatite. Of the third group at the Leswa section, four gneisses yield Proterozoic apatite U-Pb ages implying that they have not been affected by Himalayan tectonothermal events (Fig. 1.5E; Appendix Fig. A1). These old U-Pb apatite ages suggest that the rocks of the Lesser Himalayas (between MBT and MCT) have experienced a total exhumation of less than ~14 km but certainly more than 5 km (from AHe, AFT and ZHe data).

All thermochronometers with closure temperature lower than the U-Pb system (i.e. AHe, AFT and ZHe) were completely reset during Himalayan tectonics and indicate that cooling through the ZHe partial retention zone (200 °C to 140 °C) occurred between 17 and 9 Ma, with no significant difference between the two transects. The similarity of the AHe and AFT ages in most samples, and the absence of a clear age increase with elevation, suggests that the region underwent rapid exhumation of the upper few km of crust in the Pliocene. Assuming that the average ZHe age (~11 Ma) reflects cooling through 180 °C, and the average AHe age (~4 Ma) reflects cooling through 50 °C then the cooling rate for the region is approximately 19 °C/Myr implying a rate of exhumation of >0.6 mm/yr (for geothermal gradient of 30 °C/km after Wilke et al., 2012).

20

Turab, S.A., 2016

Using average cooling ages and assuming closure temperatures is a rather simplistic procedure for calculating the cooling rate. A more refined Mio-Pliocene cooling history of the region has been derived by modelling the combined datasets using QTQt (Gallagher, 2012). For the Kel profile in the structurally highest rocks, five samples (K1, K2, K3, K6 and K10; Tables 1.2, 1.3, 1.4) from which we obtained AHe, AFT and ZHe data were used for thermal history modelling. Because the vertical separation of the Kel profile samples is more than 2,000 m (Table 1.1), a palaeo-temperature offset between the top and bottom samples of 60 ± 30 °C was used in the modelling. The expected model thermal history (i.e. the weighted mean model) of the Kel profile line (Fig. 1.7) shows that the region experienced a complicated cooling history that we link to the tectonic history. Initially, the Kel region rocks appears to have cooled until about 19 Ma with the upper sample cooling to about ~120 °C at this time. This records the exhumation and cooling of rocks as a result of movement along MCT and/or GSZ that probably initiated as early as 40 Ma based on the apatite U-Pb ages. The samples then experienced reheating of about 60 °C until ~10 Ma. This was followed by a rapid cooling event between 10 and 8 Ma with a cooling rate of ~100 °C/Myr. Subsequently, slow cooling to surface temperatures took place that may have involved a period of reheating between 8 and 6 Ma.

For the Leswa profile near the MBT, the AHe, AFT and ZHe data from six samples (N8, L2, L6, L7, L9 and L10; Tables 1.2, 1.3, 1.4) were used for thermal history modelling. As this sample set spans the widely acknowledged trace of MCT (e.g. Greco and Spencer, 1993), we have modelled the samples from either side of this structure separately. This yielded results that are similar to modelling all the samples together and importantly provides additional evidence that all the samples across the trace of the MCT have observed the same thermal history as a result of exhumation along some other structure to the southwest, most probably along the MBT. The ~1,350 m of vertical separation between the top and bottom samples required a 40 ± 20 °C paleo-temperature offset. Interestingly, the modelling results for the Leswa profile show a similar thermal evolution to the Kel profile, particularly since ~10 Ma (the formation age of MBT). The Leswa region experienced a rapid cooling event at around 9 Ma where the top samples were cooled from ~180 to 60 °C in less than 1 Myr (Fig. 1.7). Samples were then reheated by about 60 °C from ~8 Ma to ~3 Ma which was followed by a second short duration (~1 Myr) rapid cooling event that brought rocks from 120-100 °C to the surface temperatures at ~2 Ma.

21

Turab, S.A., 2016

Figure 1.7: Thermal history modelling of thermochronological data for the Kel and Leswa sections. For clarity, thermal histories are shown for top and bottom samples (thick black lines) with the grey region around showing the 95% credible interval (the Bayesian form of confidence intervals) for top and bottom samples combined together. Closure temperatures of AHe, AFT and ZHe systems are also shown with dotted lines at 60 °C, 120 °C and 180 °C respectively (see also Appendix Fig. A2).

Tectonic Interpretation

This new thermochronological and geomorphic dataset allows us to propose new constraints on the history of uplift and exhumation in the Neelum river region and the evolution of the western Himalayan syntaxis region in general.

The first constraint on the tectonothermal evolution is given by apatite U-Pb ages that indicate exhumation through the ~450 °C isotherm (Chamberlain and Bowring, 2000). These data show that there are regions which are not affected by Himalayan tectonothermal events while other samples have been partly- or fully reset by Himalayan tectonics at < 50 Ma (Fig. 1.5E; Table 1.2). The boundary between the Himalayan apatite U-Pb ages and older non-reset samples lies between locations of sample L2 and L6 along the Leswa road. This is likely to represent the locus of MCT that was active since around 40 Ma, exhuming the hanging wall 22

Turab, S.A., 2016 through the 450 °C isotherm. This location confirms the position of the MCT as suggested by Greco and Spencer (1993; Figs. 1.1, 1.2). However MCT remains inactive since at least early- middle Miocene as evident from thermal history modelling of samples from both sides of its trace. The partially reset U-Pb ages record the combined effects of thrusting initially along MCT and later along MBT, and thus are partly affected by Himalayan exhumation because they were most probably exhumed from crust that was close to 450 °C 40 million years ago (Fig. 1.8). The U-Pb apatite ages that are completely reset from along Kel profile are the result of greater amounts of exhumation, the combined effect of thrusting along the GSZ, MCT and MBT (Fig. 1.8). A recent U-Pb detrital zircon provenance study (Ding et al., 2016) also confirmed enhanced Himalayan exhumation during 35-23 Ma.

Following this early stage of exhumation, thermal modelling shows that samples were at 180 - 100 °C at about 19 Ma. We suggest that the cooling during this period from ~40 to 19 Ma indicates continuous erosion following the first phases of movement on MCT and GSZ so that most of the topography was eroded away (Fig. 1.8) as evident by halt in cooling (erosion) around 19 Ma (Fig. 1.7). Samples then record reheating of about 60 °C from 19 to 10 Ma. During this period we infer reduced erosion due to the cessation or reduction of the movement along MCT and GSZ. Rapid cooling at 10-8 Ma coincides with the formation of MBT (i.e. >10 Ma; Meigs et al., 1995), and likely records the onset of movement and formation of the MBT. This cooling event is also evident from the thermal history modelling results of Leswa profile (Fig. 1.7). The north-to-south propagation of deformation also supports the MBT as the dominant source of tectonic movements at around 10 Ma. This is consistent with evidence that suggests the MCT had ceased to be active before then (Vannay et al., 2004). Our modelling results also show that samples that span the trace of the MCT have a similar cooling history from sometime between 40 Ma (based on the U-Pb apatite data) and 11 Ma (based on the ZHe data) onwards. Treloar et al. (1992) suggested that the SW-verging Kashmir thrust system (which should include both the MCT and MBT) at the eastern limb of Hazara-Kashmir syntaxis (HKS) (SW parts of our study region) is actively propagating towards the SW with its frontal thrusts cutting up through the SSE-verging Pakistan thrust system at the western limb of HKS. This also supports MBT to be the locus of main tectonic activity around 10 Ma. They further mentioned that Kashmir thrust system may overthrust and incorporate the Pakistan thrust system. This is corroborated by the 2005 Kashmir earthquake which ruptured the SW-verging Muzaffarabad thrust and hence part of Kashmir thrust system, from Balakot on the western limb of the HKS to Sudhan Gali in the axial zone of the HKS (Sayab and Khan, 2010; Turab, 2013). Formation of MBT was followed by nearly 23

Turab, S.A., 2016 isothermal to slow reheating conditions (Fig. 1.7). The final cooling to the surface temperatures may be due to incision of the Neelum river catchment into an existing topographic plateau for the Kel samples, whereas the Leswa samples show a second rapid cooling event at ~3 Ma (Fig. 1.7), possibly due to thrusting along MBT.

Figure 1.8: Schematic diagram showing five time slices of the interpreted uplift and erosion history of the Neelum river region along a schematic southwest - northeast section more or less along the Neelum river. Note that different symbols convey graphic representation of sample locations; star = Kel, square = Neelum + Leswa (E of MCT), circle = Leswa (W of MCT), black = current location, grey = location at previous time slice. 60 °C, 120 °C and 450 °C isotherms are also shown as red dashed lines. 24

Turab, S.A., 2016

This interpretation may well be supported by high ksn values in the hanging wall of MBT in Neelum valley but the sharp lithological, and thus rock strength, contrast from sandstones and shales to granites, quartzites and gneisses across its trace is also a candidate for this change in steepness index along Neelum river main channel (Fig. 1.2). We interpret the low ksn parts in the head waters of almost all the small streams as areas where the incision has not yet propagated to a recently uplifted region. Conversely, the high ksn in the main channel suggests that the entire Neelum river catchment may be the product of a young river piracy event by the Jhelum river, possibly due to the combined effect of formation of Kashmir Basin (Alam et al., 2015) and large scale drainage reorganization in the western Himalayas when all the rivers, flowing in current Punjab region changed their flow from flowing east into Ganges basin to start flowing west into Indus river basin (Clift and Blusztajn, 2005). Both drainage reorganization and the formation of Kashmir basin are reported to have taken place at ~5 Ma (Alam et al., 2015; Clift and Blusztajn, 2005). Our interpretation is also consistent with the slight AFT age increase towards the head waters of the catchment.

Cause of Exhumation

The AFT data presented above can be used together with previously published ages of the Nanga Parbat region to make a substantial improvement on the interpretation of the nature of exhumation around the western syntaxis. According to Zeitler (1985), contours of AFT age are continuous across the syntaxis and perpendicular to the major structures in the region, with base level AFT ages in the Neelum valley region being predicted to be >5 Ma (Fig. 1.1C). For some tens of kilometers around Nanga Parbat itself, the topology of these contours remains consistent with the newer data from Treloar et al. (2000), van der Beek et al. (2009) and Wilke et al. (2012) as shown on Fig. 1.1C and with other cooling age data (see e.g. Schneider et al., 2001). From this data, Zeitler (1985) inferred a region of extreme exhumation only in the region at and north of Nanga Parbat itself (see also Koons et al., 2013).

In order to widen the interpretation of the published cooling ages to the larger syntaxis region and remain consistent with previous logic, we follow the method of Zeitler (1985) and show only the base level AFT ages on Fig. 1.9. These samples are the low altitude samples K10, N5, N7 and N8 (yellow dots on Fig. 1.9; Table 1.2) with ages ranging from 2.9 ± 0.3 to 3.9 ± 0.4 Ma. Given that samples from higher elevations are not substantially older (Fig. 1.3), the

25

Turab, S.A., 2016 interpretation is robust. Combining our new data with published cooling ages shows that the ages are best connected by extending age contours from the Nanga Parbat region southward and eastwards paralleling roughly the major thrust structures around the syntaxis region. It appears that the contours drawn by Zeitler (1985) west of the syntaxis are likely to make a sharp bend southwards at the MBT. As such, the contours of exhumation age through the apatite partial annealing zone run more or less parallel to the main faults (MBT, MCT) (Fig. 1.9) with the youngest ages being located near the MCT and MBT. In this context, the “tectonic aneurysm” model of Zeitler et al. (2001a,b), clearly documented with the distribution of higher temperature cooling ages and the occurrence of low pressure granulite (Koons et al., 2013), may only be confined to Nanga Parbat itself and not include regions of the Neelum valley and south of it.

Figure 1.9: Map showing previously published apatite fission track ages plus those derived for this study (yellow dots). Following the logic of Zeitler (1985), we have selected, from our own data, only base level samples from the lowest points of the topography along the Neelum river. As such, they are comparable to the data of Zeitler allowing a re-interpretation of the contours from Fig. 1.1C.

The fact that the low-temperature thermochronology contours are parallel to the major thrusts in the western syntaxis area supports an exhumation model that advocates tectonics as the principal driving mechanism for exhumation (Fig. 1.9). This is also supported by the slight trend in AFT ages from our base level samples which increase up river, along with the ranges

26

Turab, S.A., 2016 of AFT ages from the two profile lines, Kel and Leswa. The Leswa profile is close to the main faults in the downstream parts of Neelum river and has the youngest AFT ages, whereas the Kel profile which is further from the regional faults shows somewhat older AFT ages (Table 1.2; Fig. 1.3). However the final stage of exhumation is only related to the incision of the Neelum river catchment into a plateau formed around 10 Ma due to activity along the MBT (Fig. 1.8). This interpretation is supported by stream power analysis (Fig. 1.6) although the signal from the AHe ages is not clear. Nevertheless, it is consistent with the interpretation of van der Beek et al. (2009) who argued that the Deosai Plateau in the northeastern part of the region is a remnant of the Tibetan plateau. As such, our interpretation may be of tectonic importance to a much larger region than that investigated here.

CONCLUSIONS

The results from a multiple thermochronometric dating and thermal modelling approach combined with geomorphic stream power analysis allow us to conclude the following:

Apatite U-Pb data allow to constrain the position of the MCT in the Neelum river region which has been the subject of controversy. Samples with Proterozoic U-Pb ages all are part of the Leswa profile. The boundary between the non-reset samples and all other samples affected by Himalayan tectonics (be it partly or wholly) is interpreted to represent the trace of MCT along the Leswa profile.

Average ZHe, AFT and average AHe ages are generally 9 – 16 Ma, 3 - 7 Ma and 3 – 6 Ma, respectively (Fig. 1.4). Thus the two apatite low-temperature thermochronology systems are largely indistinguishable from each other for majority of the samples. They indicate exhumation rates of 1 mm/year to 0.3 mm/year, with only a short residence time in the apatite partial annealing zone (PAZ) and AHe partial retention zone (PRZ).

Thermal history modelling of our data senses the MBT thrusting around 10 Ma in both Kel and Leswa sections (Fig. 1.7). MCT is also confirmed to have been inactive since at-least 11- 10 Ma as samples across MCT trace show same thermal history for the last 11-10 Myr even when modelled as separate groups of hanging wall and footwall samples.

0.9 The Neelum River is characterized by a steepness index of ksn > 500 m but that both the lowest reaches (where it joins the Jhelum River) as well as the headwaters of the catchment 0.9 have significantly lower steepness indices (ksn < 500 m ). We interpret this as evidence for 27

Turab, S.A., 2016 strong geomorphic disequilibrium caused by recent activity along the MBT/MCT and continuous catchment capture in the headwaters. Knick points have been identified along the channel profiles of Kunhar and Neelum rivers at a distance of ~90 and ~110 km respectively upstream of their junction with Jhelum River (Fig. 1.6). This indicates a wave of erosion propagating upstream that may have started as a result of base-level fall at 5-3 Ma.

In combination with fission track ages from the literature, our data allow well defined contouring of exhumation ages around the entire western syntaxis region. Age contours do not close in the south of Nanga Parbat, but extend towards the south-east, trending parallel to regional-scale thrust faults (Fig. 1.9), As such, exhumation is likely to be controlled by the crustal-scale thrust rather than fluvial erosion.

28

Turab, S.A., 2016

Chapter 2.

Escarpment evolution at the Red Sea continental margin of southwestern Saudi Arabia

29

Turab, S.A., 2016

ABSTRACT

Rifting of the Red Sea started around ~30-32 Ma and resulted in the formation of one of the youngest and best developed escarpments of the world: the Great Escarpment of southwestern Saudi Arabia. The escarpment is perfectly developed over a length of more than 500 km and includes mountains up to 3000 m in elevation. To better understand the geodynamics of Red Sea rifting and to constrain a denudational model for the Great Escarpment, the results of fission track and (U-Th-[Sm])/He thermochronologic techniques on apatite are combined with stream power analysis of the central part of this region. Pooled fission track ages (recording cooling through about 110 °C) range from 13.2 ± 1.7 to 352.1 ± 17.6 Ma (1σ) with all ages that are younger than about 50 Ma (and thus related to the rifting) being from elevations lower than about 500 m at the base of the escarpment. Apatite He ages range from 2.8 ± 0.3 to 264.5 ± 19.6 Ma with a similar age-elevation relationship. Bottom of the pre uplift partial annealing zone is interpreted to be lying at ~200 m present-day elevation. Our fission track data indicate that the amount of exhumation is insufficient to completely reset all the coastal plain samples, but exhumation along the escarpment appears to increase from south towards north. Highest amount of exhumation is confined to two separate regions, one in the north and second in the south, which are separated by a region of non-reset AFT ages and hence lower amount of exhumation. This interpretation is also supported by stream power analysis of the region. Reset AFT ages indicate about 4.5 km of exhumation which may have started in early Miocene but the majority of which occurred after 13.2 Ma. This interpretation is consistent with a single isolated outcrop of Nubian sandstone at the summit of Saudi Arabia’s highest peak. Distributions of AFT and AHe ages across the escarpment and coastal region supports the escarpment development by the established “downwearing” or “plateau degradation” model of escarpment evolution, which implies that the present drainage divide may have been in its position already in the Miocene. Red Sea rifting may have resulted from the complex interaction of both active and passive rifting models as it did not follow the suggested sequence of events by either of the two rifting models.

30

Turab, S.A., 2016

INTRODUCTION

In the attempts to understand the tectonic development of passive margins, the study of the morphological evolution of Great Escarpments has played an important role (Summerfield, 1989, 2000). Such studies have recognized that escarpments may be classified according to several characteristics that reflect on the tectonic processes that formed them. Based on morphological characteristics, "Arch" and "Shoulder" type margins have been discerned (Blenkinsop and Moore, 2013; Matmon et al., 2002) and these morphologies have been interpreted in terms of syn-rift surface uplift and post-rift flexural downwarp processes, respectively (Matmon et al., 2002). However, many Great Escarpments today are hundred or two hundred kilometers inland from the original rift that may have occurred tens or hundreds of million years ago, so that their original shape can only be inferred if their denudational history is understood. Therefore, based on their denudational history, Great Escarpments have been described with "Downwearing", "Downwarped" and "Escarpment Retreat" models (Persano et al., 2002; Balestrieri et al., 2005; Blenkinsop and Moore, 2013; Kooi and Beaumont, 1994; Gilchrist et al., 1994, Stüwe, 1991), each of which has different characteristics with respect to their erosional history (Fig. 2.1). Denudation in the downwarp model is almost negligible (zero/minimum) close to the coast, and increases to a maximum at the base of the present-day escarpment (Persano et al., 2002). In the escarpment retreat model, erosion is intense at the base of the escarpment throughout the history of its evolution (Fig. 2.1). Whereas in the downwearing model, the area seaward of the escarpment observes almost the same denudational rates (Balestrieri et al., 2005). In both escarpment retreat and downwearing models, greatest denudation is observed near the coast where the break-up topography and flexural rebound are the highest (Persano et al., 2002) (Fig. 2.1). Discriminating between these models is an important aid in constraining the tectonic processes that formed the margin. For example, for the 'classic' Great Escarpments of South Africa and Southeastern Australia, scarp retreat and downwearing models have been suggested respectively, thus explaining the observed flexural swell in South Africa (Stüwe, 1991) or the short time scale of overall denudation in Australia (Persano et al., 2002). Curiously, one of the most spectacular escarpments worldwide, the Great Escarpment of Saudi Arabia (Fig. 2.2) has barely been studied (with the notable exception of Bohannon et al., 1989) and its evolutionary model not been yet elucidated. This is inasmuch interesting, as the escarpment features spectacular morphological characteristics: It is up to 3000 m high (including Saudi Arabia’s highest peak: Mount Al Soudah, 2998 m), has a near vertical drop

31

Turab, S.A., 2016 of more than 1500 m and its shape remains almost unchanged for almost 1000 km (Fig. 2.3). It is flanked by the Red Sea rift, which is arguably the best understood rift worldwide. As such, the Saudi Arabian escarpment may be called a text book example of Great Escarpments and well deserves a study of its denudation history that can be related to the rift evolution. Moreover, the flanks of the Red Sea rift feature distinct morphological asymmetry between the eastern (Saudi Arabian) and the western (Egypt, Sudan) margin, and the exclusive existence of rift related volcanics (Fig. 2.2B) on the eastern side of the rift indicate a potential asymmetry in the Red Sea rift that has only been discussed based on kinematic observations (Voggenreiter and Hötzl, 1989).

Figure 2.1: Three models of escarpment migration and erosion. (A) Block diagram (3-D) view showing the initial (map view) and final/present-day configuration (bold line in cross sectional view) of escarpment formation and migration. Note that in each case, drainage pattern and drainage divide has different form and position, respectively (after Balestrieri et al., 2005). (B) Distribution of AFT age predictions across the coastal plain, escarpment and inland plateau for the three models (After Gallagher et al., 1998). (C) Distribution of AHe age predictions across the coastal plain, escarpment and inland plateau for escarpment retreat and plateau degradation models (After Braun and van der Beek, 2004).

In this paper, we present the results of a study intended to constrain the denudational model for Saudi Arabian Great Escarpment to better understand the geodynamics of Red Sea rifting in particular and continental rifting in general. For that reason, results of fission track and (U- Th-Sm)/He thermochronologic techniques on Apatite are combined with stream power analysis of the southwestern Saudi Arabian escarpment bordering the Red Sea. Apatite is a 32

Turab, S.A., 2016 powerful low-temperature thermochronometer as apatite (U-Th-Sm)/He (also termed AHe) and fission track (also termed AFT) dating records cooling of rocks through 100-40 °C (Shuster et al., 2006) and 110– 60 °C (Gleadow et al., 1983; Laslett et al., 1987), respectively. Therefore, high elevation southwestern escarpment of Saudi Arabia between Jeddah and Jizan, and its coastal plain, is sampled for this purpose (Fig. 2.3).

Figure 2.2: The Great Saudi Arabian Escarpment between Jeddah and Jizan in southwestern Saudi Arabia. (A) Satellite image of the Arabian-Nubian Shield rocks across the Red Sea. GOS: Gulf of Suez, GOA: Gulf of Aqaba, D: Djibouti, I: Israel, K: Kuwait, Q: Qatar, S: Somalia, UAE: United Arab Emirates, Jd: Jeddah, Jz: Jizan. Source: Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, and the GIS User Community. (B) Geological map of the southwestern Saudi Arabia showing the distribution of Arabian Shield rocks in the escarpment and coastal areas. Red polygons show the distribution of rift related volcanic rocks (from Brown et al., 1989). Small outcrop of sediments at Mount Al Soudah and the inland boundary between the Precambrian aged Arabian shield and Arabian platform (sedimentary) rocks is also shown. Dotted line shows location of the escarpment. Location of the spreading axis is also shown here. 33

Turab, S.A., 2016

Figure 2.3: Topography and swath profile across the southwestern Saudi Arabian escarpment with location of samples collected for this study. Samples are grouped into four roughly escarpment perpendicular transects (Lines 1 - 4). Blue line shows the drainage divide which is coincident with the escarpment lip for most of its length and red polygon shows the extent of the swath profile with half width of 100 km along the drainage divide (Prepared using Stefan Hergarten’s online tools at http://hergarten.at/; method described in Hergarten et al., 2014). For comparison of the shape of the escarpment, Swath profiles across the southeast Australian and Drakensberg escarpment of South Africa are also provided with half widths of 100 km and 150 km respectively.

34

Turab, S.A., 2016

GEOLOGICAL BACKGROUND

The Saudi Arabian Great Escarpment forms the western margin of the Arabian plate. The Arabian Plate was created ~25 Ma ago when NE Africa (current Arabia) was detached from Africa resulting in the formation of the Red Sea escarpment of southwestern Saudi Arabia (Stern and Johnson, 2010). The location of the Afar plume was controlling factor for the position of Red Sea rifting, but the direction of rifting (i.e. N30°W) was mainly due to stresses generated by slab-pull in the subduction zone in the northern Neotethys beneath the current Bitlis-Zagros thrust zone, towards the northeast (Bosworth et al., 2005). Direction of the Red Sea extension changed at ~14 Ma from rift normal (N60°E) to highly oblique and parallel to the Aqaba–Levant transform (N15°E) possibly due to the waning forces of the Afar plume, for which there is little evidence after about 25 Ma (Bosworth et al., 2005). The timing of the change of direction in the rifting process also corresponds to the collision of Arabia with Eurasia and creation of a transform boundary cutting through Sinai (Egypt) and the Levant continental margin, linking the northern Red Sea with the Bitlis-Zagros convergence zone (Bosworth et al., 2005).

Although details of the escarpment evolution are not well known, it is clear that the rift flank uplift associated with the rifting processes caused widespread denudation, removing the entire sequence of Mesozoic sedimentary cover (Nubian sandstone) that covered much of northern Africa at the end of the Mesozoic. This sequence was more than 2000 m thick in the northern Red Sea and Sinai and gets progressively thinner to the south, being of the order of 500 m thick in the region of interest (Garfunkel, 1988). Today these sediments are still exposed in large parts of Egypt west of the Eastern Desert and in northern Sinai. In Saudi Arabia, these sediments commence some 200 km east of the coast and continue across the Arabian peninsula. Interestingly, a single outcrop, a mere few square kilometers in size, is preserved at the very highest point of the Saudi Arabian Escarpment at Mount Al Soudah (Figs. 2.2B, 2.4B) placing a tight constraint on the overall amount of exhumation related to the escarpment evolution.

The escarpment evolution is also constrained by young volcanism that covers large areas of the coastal plains below the escarpments. These volcanics are some 2-6 Ma old and are known by the local names of Al Birk, Tuffil (Shama), Rahat, and Lunayyir volcanic fields (Fig. 2.2B; Brown et al., 1989). These volcanics lie directly over rocks of Arabian Shield which are exposed along the (north)eastern escarpment of Red Sea rifting and inland for hundreds of kilometers (Figs. 2.2A,B) and which were the continuation of Nubian Shield 35

Turab, S.A., 2016 rocks exposed on the (south)western side of Red Sea. The Arabian Shield is a stable craton that remained uplifted since Cambrian times and is composed primarily of late Precambrian aged rocks including about 50% plutonic rocks and 50% volcanic and sedimentary rocks where among the plutonic rocks, about 70% are granitic in composition (Gettings et al., 1986; Stern and Johnson, 2010) thus are appropriate for AHe and AFT analysis.

Figure 2.4: (A) Near vertical drop at the escarpment lip. For scale, arrows show two men standing. (B) Outcrop of Mesozoic sediments (Nubian sandstone) lying on top of the crystalline rocks of Arabian shield near the highest point of Saudi Arabian escarpment (Mount Al Soudah). (C) View from the escarpment lip towards the high elevation, low relief inland plateau. (D) Google-Earth view of the escarpment region showing intense erosion at escarpment face indicated by sharp, pointed interfluves with almost flat lying landscape towards the inland plateau. Red line shows drainage divide from Figure 2.3 and is coincident with the escarpment lip.

Morphologically, the Saudi Arabian escarpment follows the entire Red Sea coast, However, it is only south of Jeddah that it reaches elevations above 2000 m and there follows about 500 km to the south into Yemen. The inland side of the Escarpment is almost flat (with few small hills) that downslope gradually towards central Saudi Arabia and gives the distinct impression of a mature landscape (Fig. 2.4C,D). The escarpment forms the drainage divide for majority of its length in the study area with exception of few deviations of drainage divide inland for smaller regions. Extreme erosion is concentrated near the escarpment base as evident from the

36

Turab, S.A., 2016 formation of sharp pointing interfluves (Fig. 2.5). Peak elevations along the entire Red Sea escarpment in Arabia generally increases from north to south. In the north, they range about 2100-2400 m, and in the southern Saudi Arabia, near the town of Abha, the escarpment reaches to as high as ~3000 m at Mount Al Soudah, the highest point of Saudi Arabia. The escarpment continues to rise further south as it enters into Yemen and near the capital Sana’a, it gains an elevation of about ~3650 m at Mount Nabi Shu’ayb, the highest point of entire Arabian peninsula. Shape of the escarpment is a classical representation of escarpment separating high elevation, low relief plateau from low elevation coastal plains with a near vertical drop of more than 1500 m in about less than 5 km of map distance (Fig. 2.3). The inland plateau of Saudi Arabian escarpment has average elevations of 1000-1500 m (Bohannon et al., 1989). Physiography in general, i.e. the increasing trend in elevations from north to south along the escarpment and the average elevations of inland plateau and escarpment itself, is higher on Arabian as compared to the African side of Red Sea margin showing a probable inherited asymmetry of Red Sea rifting processes.

LOW TEMPERATURE THERMOCHRONOLOGY: SAMPLES AND TECHNIQUES

For young high-elevation rifted margins, it is possible to determine the timing and rates of post-break up denudation and hence the model for escarpment evolution by carrying out fission track and (U-Th-Sm)/He dating on apatite separated from samples collected along transects perpendicular and parallel to the escarpment (Balestrieri et al., 2005; Braun and van der Beek, 2004). Minimum age calculated is a controlling factor, its value gives information about the rate of escarpment migration and its location provides evidence about the migration mode (Braun and van der Beek, 2004). Hence a total of 50 samples, forming a decent geographical and altitudinal spread throughout the study area, were collected which can be divided into four roughly escarpment perpendicular profile/transect lines across the escarpment (Fig. 2.3). Based on lithology, elevation and position on the map, 22 samples were analyzed using apatite (U-Th-[Sm])/He dating out of which 18 samples were also dated by apatite fission track method (Tables 2.1, 2.2, 2.3; Figs. 2.3, 2.6A-F). 4-6 samples per profile line were selected such that each line contains at least 1 sample from top of the escarpment and rest of the samples covering maximum length of the area between escarpment and the Red Sea coast.

37

Turab, S.A., 2016

Figure 2.5: (A-C) Different views from the top of the escarpment towards the coastal plain with sharp interfluves in (C) pointing to rapid erosion (?). (D) View towards the top of the escarpment from half way up the road going towards the top.

THERMOCHRONOLOGICAL RESULTS

AFT ages range from 13.2 ± 1.7 to 352.1 ± 17.6 Ma (1σ) (Fig. 2.6A, Table 2.2). Samples 31, 34, 37, 46 had sufficient confined tracks to allow track length measurements with mean track length (MTL) of the samples ranging from 10.57 ± 1.73 to 11.59 ± 2.12 µm (Fig. 2.7). The kinetic parameter, Dpar, shows a range from 1.41 ± 0.18 to 1.88 ± 0.26 µm, while the chlorine 38

Turab, S.A., 2016 content ranges from 0.00 (-0.06 exact value) to 0.19 wt % (Table 2.2). All samples but two (22 and 37) passed the chi-squared (χ2) test (Galbraith, 1981, modified for LA-ICP-MS fission track dating by Donelick et al., 2005).

TABLE 2.1. DETAILS OF SAMPLES TAKEN FOR LOW TEMPERATURE THERMOCHRONOLOGIC ANALYSIS Sample Latitude* Longitude* Sample Elevation Rock type (°N) (°E) reference (m) 3 20 34.105 40 28.014 Line 1 260 Granite 4 20 27.684 40 28.272 Line 1 167 Granite 6 19 45.488 40 56.209 Line 2 67 Granite 8 19 53.616 41 03.227 Line 2 180 Granite 9 19 54.974 41 08.598 Line 2 248 Granite 14 19 59.800 41 26.369 Line 2 1273 Magma mingling (granitic + mafic rock) 22 20 43.793 40 49.424 Line 1 2277 Granite 24 20 58.250 41 04.615 Line 1 1494 Granodiorite-diorite 27 20 01.878 41 26.980 Line 2 2211 Granite 31 19 06.561 42 07.759 Line 3 2398 - 34 18 13.255 42 31.333 Line 4 2221 Granite 37 18 16.771 42 19.945 Line 4 1598 Dark greyish rock with some light colored minerals 38 17 53.674 42 15.069 Line 4 169 Pegmatite/Granite intruding foliated metamorphic rock of staurolite grade 39 17 52.176 42 12.516 Line 4 116 Aplite, with little or no biotite 40 18 07.273 41 38.745 Line 4 40 Schist + granite 42 19 05.828 41 33.207 Line 3 147 Granite 43 19 06.269 41 38.100 Line 3 212 Granite 44 19 04.715 41 47.607 Line 3 289 Granite 45 18 47.624 41 59.198 Line 3 393 Granite, tonalite 46 19 07.219 42 03.857 Line 3 875 Granite 49 19 42.773 41 24.036 Line 2 313 Granite 50 20 25.098 39 58.410 Line 1 42 Granite *data entered in DDM (Degrees Decimal Minutes) format.

Single grain AHe ages of our data set range from 1.88 ± 0.35 to 207.87 ± 15.42 Ma (1σ, uncorrected) and from 2.82 ± 0.33 to 264.46 ± 19.62 Ma (1σ, FT-corrected) (Table 2.3). Average apatite He ages (Ave. AHe) typically are younger than or overlap the corresponding AFT ages (Figs. 2.6B-F). Both AFT and AHe ages show a trend of increasing ages with elevation for all profiles with exception of few samples (Figs. 2.6B-F; Table 2.3). As all the samples are grouped into four roughly escarpment perpendicular profile lines, hence we describe the data in details from each profile line separately; 39

Turab, S.A., 2016

Line 1

Line 1 is the northwestern profile line of the study area having 5 samples in total (3, 4, 22, 24 and 50, Fig. 2.6B). Single grain AHe ages of Line 1 samples ranges from 1.88 ± 0.35 to 106.63 ± 5.09 Ma (Uncorrected) and 2.82 ± 0.33 to 183.33 ± 6.3 Ma (Corrected) whereas AFT ages are 13.2 ± 1.7 to 138.3 ± 8.7 Ma (Tables 2.2, 2.3). Average AHe ages for these samples range from 3.6 ± 1.4 to 109.6 ± 22.8 Ma. Generally, the high elevation samples show old AFT and AHe ages with high spread of single grain AHe ages (Fig. 2.6B) whereas the low elevation samples display younger AFT and more or less similar (single grain) AHe ages except sample 50 which shows high spread of AHe ages. Reason for this could be the poor quality of Apatite crystals. It is important to mention here that sample 24 is from inland plateau ~35 km towards the mainland from the escarpment tip (Fig. 2.3).

Line 2

There are six samples analyzed from Line 2 (6, 8, 9, 14, 27 and 49), which gave single grain AHe ages of 1.96 ± 0.31 to 176.82 ± 4.39 Ma (Uncorrected) and 3.13 ± 0.5 to 252.24 ± 6.26 Ma (Corrected) while AFT ages of the samples are 17.8 ± 1.9 to 177.4 ± 12.7 Ma (Tables 2.2, 2.3). The average AHe ages of Line 2 samples show a range of 3.8 ± 0.7 to 155.0 ± 60.0 Ma. As with Line 1 samples, Line 2 samples also show positive correlation of the sample elevations with AFT and AHe ages with the exception of sample 49 which display high spread of single grain AHe ages (Fig. 2.6C). Sample 27, being at high elevation (on top of escarpment), is showing older ages with large spread of single grain AHe ages.

40

Turab, S.A., 2016

TABLE 2.2: APATITE FISSION TRACK RESULTS

Sample Number Ns ΣpiΩi 1σΣpiΩi χ2 P(χ2) Cl 1σ Cl Dpar 1σ Dpar Pooled age Elevation of grains wt% wt% (μm) (μm) (Ma ± 1σ) (m) 3 26 474 6.35E-05 5.65E-07 20.03 0.75 0.05 0.16 1.73 0.18 73.5 ± 3.7 260 4 26 59 4.43E-05 5.81E-07 29.45 0.25 0.12 0.13 1.41 0.18 13.2 ± 1.7 167 6 26 91 5.05E-05 8.27E-07 25.40 0.44 0.16 0.17 1.53 0.2 17.8 ± 1.9 67 8 28 90 4.52E-05 1.65E-06 22.15 0.73 0.01 0.16 1.52 0.17 19.7 ± 2.2 180 22 29 443 4.55E-05 3.42E-07 111.38 0.00 0.02 0.16 1.57 0.16 95.8 ± 5.0 2277 24 27 291 2.06E-05 2.73E-07 26.43 0.44 0.00 0.15 1.53 0.19 138.3 ± 8.7 1494 27 26 222 1.22E-05 1.85E-07 31.58 0.17 0.04 0.22 1.45 0.24 177.4 ± 12.7 2211 31 29 518 1.42E-05 1.97E-07 37.78 0.10 0.02 0.24 1.88 0.26 352.1 ± 17.6 2398 34 26 1063 3.11E-05 2.36E-07 26.13 0.40 0.01 0.16 1.69 0.21 330.6 ± 12.3 2221 37 25 801 1.09E-04 1.07E-06 2202.56 0.00 0.08 0.12 1.71 0.18 72.3 ± 3.0 1598 38 28 176 8.02E-05 5.84E-07 18.46 0.89 -0.06 0.12 1.54 0.21 21.7 ± 1.7 169 39 18 36 1.27E-06 2.51E-08 16.16 0.51 0.01 0.17 1.44 0.19 274.8 ± 46.4 116 40 21 75 2.60E-06 3.75E-08 23.85 0.25 0.16 0.30 1.52 0.18 280.1 ± 33.1 40 43 30 523 6.75E-05 5.01E-07 28.48 0.49 0.02 0.13 1.54 0.23 76.3 ± 3.7 212 45 26 188 4.07E-05 7.83E-07 29.80 0.23 0.00 0.14 1.59 0.19 45.6 ± 3.6 393 46 28 824 6.71E-05 4.69E-07 31.82 0.24 0.19 0.12 1.85 0.19 120.5 ± 4.9 875 49 26 168 2.34E-05 4.03E-07 36.45 0.07 -0.02 0.09 1.72 0.21 70.7 ± 5.8 313 50 26 40 2.37E-05 4.66E-07 18.78 0.81 0.02 0.12 1.50 0.19 16.7 ± 2.7 42 Note: Details about methodology and symbols are given in Supplementary material file and references therein. Zeta calibration value (ζ) is 19.82 ± 0.39.

41

Turab, S.A., 2016

Figure 2.6: Low temperature thermochronological results and its variation with sample elevation. Red dashed line shows the timing of Red Sea rift initiation (~30-32 Ma). (A) AFT ages of all the samples with bottom of the exhumed-, paleo-PAZ interpreted at ~200 m present day elevation (see text for details). (B, C, E, F) AHe, AFT and average AHe ages from four roughly escarpment perpendicular transects (named ‘Lines’ here) plotted against sample elevations (see text for details). (D) For comparison, AFT data from this study is plotted along previously published AFT data from the same region by Bohannon et al. (1989).

42

Turab, S.A., 2016

TABLE 2.3: APATITE (U-Th-[Sm])/He RESULTS Sample Mass U Th Sm 4He eU Raw Error FT Corrected Average age Th/U Error Diameter (µg) (ppm) (ppm) (ppm) (nmol/g) (ppm) Age Age (Ma) (µm) (Ma) (Ma) 3a 1.31 11.85 22.25 235.94 1.19 17.08 12.697 0.333 0.65 19.5 ± 0.5 1.89 0.08 73.8

3b 3.99 12.97 25.34 248.58 1.47 18.92 14.126 0.309 0.76 18.6 ± 0.4 1.97 0.07 114.1

3c 1.92 14.20 27.20 202.79 2.11 20.59 18.720 0.453 0.69 27.2 ± 0.7 1.93 0.07 88.1

3d 1.38 10.61 19.56 186.47 0.81 15.21 9.752 0.330 0.64 15.1 ± 0.5 1.86 0.08 69.1

3e 2.44 8.36 17.73 171.56 1.40 12.53 20.233 0.506 0.71 28.4 ± 0.7 2.14 0.08 79.9

3f 1.36 9.37 16.48 N.D.* 0.68 13.24 9.443 0.688 0.65 14.6 ± 1.1 1.77 0.30 71.1

3g 2.42 9.70 17.37 N.D.* 0.79 13.78 10.657 0.452 0.70 15.3 ± 0.6 19.8 ± 5.8 1.80 0.17 84.4 4a 4.29 3.28 13.23 678.44 0.09 6.39 2.298 0.092 0.81 2.8 ± 0.1 4.06 0.16 115.9

4b 1.80 2.55 6.81 427.82 0.05 4.15 1.884 0.218 0.67 2.8 ± 0.3 2.69 0.25 77.6

4c 1.37 3.47 9.88 366.15 0.13 5.79 3.953 0.243 0.70 5.7 ± 0.3 2.87 0.23 70.1

4d 0.92 2.57 9.85 533.41 0.06 4.89 1.876 0.346 0.65 2.9 ± 0.5 3.6 ± 1.4 3.85 0.65 69.3 6c 1.00 2.37 10.90 670.25 0.06 4.93 1.958 0.313 0.63 3.1 ± 0.5 4.64 0.75 62.8

6d 1.98 5.60 20.42 400.48 0.16 10.40 2.685 0.107 0.70 3.8 ± 0.2 3.67 0.17 79.9

6e 0.79 20.14 96.62 819.95 0.66 42.84 2.798 0.083 0.62 4.5 ± 0.1 3.8 ± 0.7 4.83 0.19 62.9 8a 12.94 16.08 45.00 311.64 2.76 26.65 18.859 0.399 0.82 23.0 ± 0.5 2.82 0.08 152.9

8b 2.45 4.51 12.11 168.85 0.64 7.36 15.676 0.426 0.71 22.2 ± 0.6 2.70 0.13 94.0

8c 1.24 13.73 37.86 314.42 2.87 22.63 23.047 0.698 0.63 36.5 ± 1.1 2.78 0.12 68.6

8d 2.32 32.82 57.01 555.52 2.74 46.22 10.801 0.235 0.71 15.2 ± 0.3 1.75 0.05 78.4

8e 3.24 11.13 12.41 145.71 3.30 14.05 42.773 1.013 0.73 58.5 ± 1.4 1.12 0.04 96.7

8f 4.51 5.98 6.75 N.D.* 0.62 7.57 15.181 0.709 0.75 20.1 ± 0.9 1.14 0.10 104.3

8g 2.67 7.91 11.63 N.D.* 1.44 10.64 24.927 1.558 0.71 34.9 ± 2.2 1.48 0.21 93.3

8h 1.57 27.17 52.17 N.D.* 3.61 39.43 16.893 0.463 0.65 26.2 ± 0.7 29.6 ± 13.7 1.93 0.11 68.6 9a 1.50 5.78 21.24 N.D.* 0.58 10.77 9.970 0.668 0.65 15.2 ± 1.0 3.70 0.90 66.9

9b 1.17 11.63 9.48 N.D.* 0.96 13.86 12.815 1.420 0.66 19.4 ± 2.2 0.82 0.18 66.7

9c 1.31 5.73 15.30 N.D.* 0.60 9.33 11.945 1.182 0.68 17.7 ± 1.7 2.69 0.79 65.5

43

Turab, S.A., 2016

9d 1.15 7.42 7.61 N.D.* 1.02 9.21 20.539 3.024 0.63 32.8 ± 4.8 1.03 0.30 59.0

9e 1.08 7.21 18.68 N.D.* 1.53 11.60 24.338 2.506 0.56 43.3 ± 4.5 25.7 ± 12.0 2.61 0.78 52.0 14a 1.38 16.28 35.22 N.D.* 4.04 24.55 30.390 1.324 0.68 44.4 ± 1.9 2.18 0.23 73.8

14b 0.84 22.43 48.24 N.D.* 8.48 33.77 46.276 2.341 0.63 73.9 ± 3.7 2.17 0.28 62.5

14c 0.94 13.26 53.45 N.D.* 6.18 25.82 44.103 2.344 0.69 64.3 ± 3.4 4.06 0.70 68.7

14d 2.26 20.37 61.71 N.D.* 7.11 34.88 37.574 3.578 0.71 52.9 ± 5.0 3.05 0.17 81.3

14e 5.40 23.66 62.22 N.D.* 7.45 38.28 35.899 0.657 0.77 46.8 ± 0.9 56.5 ± 12.4 2.65 0.08 118.1 22a 7.23 2.30 4.43 26.38 1.19 3.34 65.162 5.591 0.79 82.4 ± 7.1 1.94 0.38 136.7

22b 1.75 4.35 13.25 119.40 2.58 7.46 62.502 7.409 0.67 93.8 ± 11.1 3.07 1.00 80.0

22c 4.51 3.84 4.17 25.81 2.37 4.82 89.790 10.551 0.76 118.1 ± 13.9 1.10 0.26 115.0

22d 2.88 18.92 37.45 108.07 12.02 27.72 79.365 2.612 0.71 112.4 ± 3.7 1.99 0.09 85.0

22e 4.54 4.34 11.56 59.26 4.14 7.05 106.629 5.094 0.75 141.4 ± 6.8 109.6 ± 22.8 2.69 0.19 111.7 24c 2.18 1.80 4.57 93.56 0.73 2.87 45.117 9.287 0.67 67.5 ± 13.9 2.57 0.85 75.0

24d 2.53 1.26 0.85 163.74 0.59 1.46 65.241 34.671 0.69 94.1 ± 50.0 0.68 0.88 83.3

24e 3.55 2.05 1.68 103.77 0.46 2.45 33.256 7.190 0.74 44.9 ± 9.7 0.82 0.39 88.3

24f 2.24 1.85 2.63 N.D.* 0.82 2.47 61.163 17.626 0.69 89.2 ± 25.7 1.43 0.90 81.1

24g 1.91 1.56 5.66 N.D.* 0.81 2.89 51.294 10.764 0.69 74.4 ± 15.6 74.0 ± 19.5 3.64 2.84 76.4 27a 2.48 12.23 49.99 374.02 23.64 23.97 176.823 4.390 0.70 252.2 ± 6.3 4.12 0.14 92.6

27b 5.71 7.23 23.87 334.16 7.07 12.84 97.964 1.851 0.79 124.2 ± 2.3 3.32 0.10 115.7

27c 3.67 5.18 21.86 283.98 7.90 10.31 135.719 4.083 0.75 181.4 ± 5.5 4.25 0.15 91.7

27d 1.58 7.31 21.42 183.20 3.47 12.34 50.791 1.276 0.66 77.3 ± 1.9 2.95 0.14 76.5

27e 1.65 9.87 31.80 356.97 10.76 17.34 111.060 2.814 0.67 167.0 ± 4.2 3.25 0.13 80.3

27g 0.67 8.40 26.37 N.D.* 5.57 14.59 70.153 9.240 0.55 128.0 ± 16.9 155.0 ± 60.0 3.16 1.38 55.4 31a 1.96 1.89 7.21 108.03 1.81 3.59 89.480 4.098 0.73 122.7 ± 5.6 3.84 0.48 80.1

31b 1.50 2.32 8.88 156.84 2.76 4.40 110.187 5.029 0.63 176.3 ± 8.0 3.85 0.48 64.0

31d 6.96 2.08 8.73 144.81 3.31 4.13 140.624 3.209 0.80 175.3 ± 4.0 4.23 0.19 126.2

31e 1.63 1.99 12.21 192.04 3.13 4.86 112.702 4.105 0.65 172.3 ± 6.3 161.7 ± 26.0 6.18 0.83 78.5 34a 4.27 15.35 11.69 37.80 16.21 18.10 163.288 12.079 0.76 215.7 ± 16.0 0.77 0.10 108.3

34b 6.57 16.30 13.89 46.41 13.79 19.56 128.886 9.573 0.81 159.9 ± 11.9 0.86 0.11 121.7

44

Turab, S.A., 2016

34c 7.33 10.59 10.02 48.72 14.84 12.95 207.867 15.420 0.79 264.5 ± 19.6 0.95 0.13 118.3

34d 3.64 12.86 11.45 61.13 11.20 15.56 131.324 5.531 0.74 176.5 ± 7.4 0.90 0.07 103.3

34e 5.59 17.78 17.23 40.10 12.15 21.83 101.980 3.086 0.78 130.9 ± 4.0 189.5 ± 51.9 0.98 0.04 103.3 37a 0.69 7.94 6.32 6.49 1.55 9.43 30.339 2.033 0.61 49.4 ± 3.3 0.80 0.08 59.2

37b 1.58 21.99 17.18 47.95 20.77 26.02 145.790 4.524 0.68 216.0 ± 6.7 0.79 0.03 82.3

37d 4.12 15.02 13.26 57.65 17.02 18.13 170.812 4.519 0.76 225.6 ± 6.0 163.7 ± 99.1 0.89 0.03 114.8 38a 3.40 3.70 1.28 188.02 3.72 4.00 160.680 6.858 0.74 217.1 ± 9.3 0.35 0.03 99.5

38b 2.39 20.71 16.24 1210.61 3.53 24.52 25.091 0.612 0.70 35.6 ± 0.9 0.79 0.03 74.4

38c 7.92 9.94 12.61 1060.61 1.83 12.90 23.796 0.616 0.81 29.3 ± 0.8 1.28 0.05 124.4

38e 9.06 10.77 10.22 1180.81 1.77 13.17 22.309 0.514 0.81 27.5 ± 0.6 0.96 0.03 149.4

38f 3.07 23.35 23.61 N.D.* 4.03 28.90 25.764 0.755 0.76 33.9 ± 1.0 1.02 0.05 99.5

38g 6.74 1.92 0.24 N.D.* 0.37 1.98 34.212 6.314 0.82 41.7 ± 7.7 0.12 0.13 135.1

38h 3.91 18.77 56.68 N.D.* 7.52 32.09 43.205 0.900 0.76 56.6 ± 1.2 3.04 0.12 103.1

38i 2.49 7.29 5.31 N.D.* 2.20 8.53 47.657 4.201 0.71 67.2 ± 5.9 63.6 ± 63.5 0.73 0.13 76.3 39a 0.95 4.62 4.97 206.41 3.67 5.78 111.491 7.408 0.60 186.8 ± 12.4 1.08 0.14 60.4

39c 2.26 1.85 0.00 15.11 1.00 1.85 98.427 8.560 0.79 125.1 ± 10.9 0.00 #DIV/0! 98.4

39d 0.94 2.51 0.65 110.39 2.26 2.66 147.739 26.284 0.61 244.2 ± 43.4 185.3 ± 59.6 0.26 0.13 63.6 40a 1.31 1.20 4.93 127.47 0.76 2.36 55.409 4.522 0.68 81.1 ± 6.6 4.14 1.08 68.3

40b 0.92 5.32 13.99 229.00 0.68 8.61 14.030 0.593 0.58 24.1 ± 1.0 2.65 0.26 58.3

40c 0.59 2.21 7.98 150.73 0.38 4.08 16.642 1.816 0.53 31.5 ± 3.4 3.64 1.36 47.0

40d 1.61 1.16 0.00 43.48 0.50 1.16 76.581 17.089 0.71 107.7 ± 24.0 61.1 ± 40.1 0.00 #DIV/0! 83.7 42h 0.55 15.04 40.99 105.25 4.68 24.68 34.983 2.274 0.52 67.7 ± 4.4 2.74 0.54 50.9

42i 0.36 6.93 13.37 61.52 1.45 10.08 26.549 6.412 0.54 48.9 ± 11.8 1.94 1.22 46.7

42j 0.88 6.98 4.12 104.92 0.76 7.95 17.653 3.356 0.63 28.2 ± 5.4 48.3 ± 19.7 0.60 0.23 57.7 43a 1.18 12.00 0.00 104.38 2.67 12.00 40.778 2.032 0.68 59.7 ± 3.0 0.00 #DIV/0! 72.5

43b 0.63 26.32 28.82 154.47 4.73 33.09 26.263 0.691 0.54 49.1 ± 1.3 1.10 0.05 49.3

43c 1.80 2.72 6.86 156.05 0.97 4.33 39.440 2.294 0.64 61.4 ± 3.6 2.54 0.25 61.9

43d 1.73 7.48 10.20 112.15 2.25 9.87 41.375 0.987 0.67 61.9 ± 1.5 58.0 ± 6.0 1.37 0.06 76.0 44a 3.64 1.32 0.00 N.D.* 0.16 1.32 23.116 6.315 0.76 30.5 ± 8.3 0.00 #DIV/0! 95.3

45

Turab, S.A., 2016

44b 1.31 3.21 6.16 N.D.* 0.36 4.66 14.122 1.597 0.63 22.6 ± 2.6 1.93 0.99 66.1

44c 1.63 1.13 0.00 N.D.* 0.16 1.13 25.808 7.318 0.69 37.5 ± 10.6 0.00 #DIV/0! 74.7

44e 3.61 1.35 0.00 N.D.* 0.15 1.35 19.912 6.496 0.74 26.8 ± 8.7 29.3 ± 6.3 0.00 #DIV/0! 104.2 45a 2.19 3.75 4.31 36.25 3.60 4.76 137.329 5.349 0.69 197.9 ± 7.7 1.16 0.08 84.4

45c 3.00 2.31 1.16 5.08 0.29 2.58 20.671 1.224 0.74 27.8 ± 1.6 0.51 0.06 89.8

45e 1.69 7.49 8.93 15.79 5.07 9.59 96.955 3.142 0.67 144.9 ± 4.7 123.5 ± 87.0 1.20 0.05 76.6 46a 0.55 6.52 31.52 151.66 3.19 13.93 41.677 1.536 0.51 81.4 ± 3.0 4.87 0.61 50.1

46b 3.12 21.87 70.89 242.66 15.45 38.53 73.148 1.816 0.78 93.4 ± 2.3 3.26 0.11 100.3

46c 0.94 30.95 44.58 266.86 13.01 41.43 57.434 1.378 0.63 90.7 ± 2.2 1.45 0.06 61.5

46e 2.61 13.89 30.67 109.99 9.72 21.09 84.202 1.882 0.72 116.6 ± 2.6 95.5 ± 15.0 2.22 0.07 83.6 49a 1.44 4.55 10.93 355.60 4.00 7.12 97.311 2.947 0.67 145.0 ± 4.4 2.42 0.17 70.9

49b 1.39 5.57 14.30 268.53 5.20 8.93 103.061 3.440 0.66 156.6 ± 5.2 2.59 0.17 67.4

49c 1.16 7.46 27.07 507.28 3.51 13.82 44.729 1.288 0.62 72.6 ± 2.1 3.65 0.18 64.5

49d 0.56 16.86 59.88 846.56 6.91 30.93 39.817 0.954 0.58 69.2 ± 1.7 3.58 0.17 55.0

49e 1.58 7.32 17.19 384.11 3.25 11.36 50.578 1.227 0.64 78.7 ± 1.9 2.37 0.11 67.5

49f 1.84 5.54 19.19 N.D.* 1.52 10.06 27.909 2.071 0.70 40.0 ± 3.0 3.49 0.90 79.3

49g 2.38 11.22 17.79 N.D.* 13.68 15.40 162.410 8.124 0.76 215.1 ± 10.8 111.0 ± 62.4 1.60 0.17 106.2 50a 0.67 6.97 18.36 594.30 6.50 11.29 99.365 3.415 0.54 183.3 ± 6.3 2.65 0.22 52.0

50c 1.15 15.65 14.11 243.14 5.21 18.97 49.893 1.298 0.68 72.9 ± 1.9 0.91 0.04 75.6

50d 0.92 7.40 18.71 346.93 5.38 11.79 80.935 2.516 0.62 130.1 ± 4.0 2.55 0.16 61.1

50e 0.71 9.48 32.21 400.44 5.89 17.05 61.749 1.574 0.54 114.3 ± 2.9 3.42 0.20 47.7

50f 0.88 11.31 32.45 N.D.* 0.83 18.94 8.116 0.473 0.65 12.5 ± 0.7 2.89 0.52 70.0

50h 3.70 5.47 14.74 N.D.* 0.77 8.93 15.882 0.550 0.76 20.8 ± 0.7 89.0 ± 66.3 2.72 0.25 102.8 Note: Average ages and associated errors are calculated as mean and standard deviation of all the aliquots from the same sample. *N.D. = not determined

46

Turab, S.A., 2016

Figure 2.7: Track lengths from four samples. Samples 31 and 34 are from the top of the escarpment while samples 37 and 46 are from the face of the escarpment and coastal plain respectively.

Line 3

Line 3 contains six samples (31, 42, 43, 44, 45 and 46). Single grain AHe ages are from 14.12 ± 1.60 to 140.62 ± 3.21 Ma (Uncorrected) and 22.59 ± 2.56 to 197.88 ± 7.71 Ma (Corrected) with AFT ages of 45.6 ± 3.6 to 352.1 ± 17.6 Ma (Fig. 2.6E). Average AHe ages are ranging from 29.3 ± 6.3 to 161.7 ± 26.0 Ma. Overall, older AFT versus AHe ages from the same samples, good AHe single grain age reproducibility as well as increasing trend in AFT and AHe ages with increase in elevation is observed except sample 45 (Fig. 2.6E). Sample 45 lies some 35-40 km south of the general trend of Line 3 samples hence there is every chance that there has been some fault(s) between sample 45 and the rest of the Line 3 samples or there is gradual change in rate of exhumation (see later). Nevertheless, the poor age reproducibility for AHe single grain analysis and similar AHe and AFT age from this sample remains unexplained and probably is a consequence of poor quality of apatite grains.

47

Turab, S.A., 2016

Line 4

Line 4 is the southeastern profile line of our study area (Fig. 2.3) containing five samples (34, 37, 38, 39 and 40). Single grain AHe ages of Line 4 samples ranges from 14.03 ± 0.59 to 207.87 ± 15.42 Ma (Uncorrected) and 24.15 ± 1.02 to 264.46 ± 19.62 Ma (Corrected) with AFT ages of 21.7 ± 1.7 to 330.6 ± 12.3 Ma (Fig. 2.6F). Average AHe ages of these samples show a range of 61.1 ± 40.1 to 189.5 ± 51.9 Ma. Contrary to other three profile lines, Line 4 samples display poor age reproducibility of AHe single grain ages, which is evident from the large errors associated with the average AHe ages above, as well as weak trend of increasing AFT and AHe ages with the sample elevations.

MORPHOLOGICAL ANALYSIS

Stream profile analysis was carried out using 3 arc second (~90m) resolution Digital Elevation Model (DEM) data along the Saudi Arabian Escarpment region roughly between Jeddah and Jizan (Fig. 2.8). Following common procedure, we study channel characteristics in comparison to the geomorphic steady state where channel slope (S) and drainage area (A) -θ are related by a power law function S= ks A where ks and θ are referred to as steepness and concavity indices, respectively (e.g. Wobus et al., 2006). The values of ks and θ can be measured directly from DEM data by performing a regression analysis on log–log plot between A and S. The normalized steepness index (ksn) is calculated using a fix reference concavity value (θref) which is usually taken around 0.45 (Wobus et al., 2006). The ksn is proportional to erosion rate and can thus be used to infer zones of variable exhumation

(Whittaker, 2012). The ksn of all the streams was calculated in Matlab R2011b by linking it to the ArcMap 10.1 through the “Profiler toolbar” using a value of 0.45 for θref (for details see Whipple et al., 2007).

The ksn values are predominantly high in two regions of study area (Red parts of streams in Fig. 2.8), one near Abha in the south and the second near Maysaan in the northern parts. The region separating these two areas displays relatively lower ksn values (green parts of streams) whereas the more flatter segments of streams, i.e. in the coastal plain and the inland plateau, has very low ksn values (white parts of streams).

A first-hand look at the figure 2.8 give signs that southern segment of Saudi Arabian escarpment is shoulder type of escarpment as the drainage divide is coincident with the

48

Turab, S.A., 2016 escarpment (Fig. 2.3) but detailed analysis reveal that there are some smaller areas along the escarpment where the drainage divide shifts inland and streams from inland plateau flow towards the coastal plain across the escarpment (Fig. 2.8) thus making this classification little ambiguous.

Figure 2.8: Results of geomorphic analyses of the southwestern Saudi Arabian escarpment area between Jeddah and Jizan. Yellow dashed polygons show regions of high ksn and reset (< ~32 Ma) AFT ages from Bohannon et al. (1989) as well as this study. Blue line shows drainage divide from figure 2.3 which at places shifts towards inland plateau (black dashed ellipse). 49

Turab, S.A., 2016

DISCUSSION

Total Amount and Timing of Uplift

The Red Sea continental margins have observed ~4 - >5 km of uplift since the initiation of Red Sea rifting (Kohn and Eyal, 1981; Bohannon et al., 1989; Steckler and Omar, 1994; Bojar et al., 2002; Balestrieri et al., 2005). In central and southern Red Sea escarpment region of Saudi Arabia, Bohannon et al. (1989) studied almost the same area as studied here and reported about 4 km of uplift since 20 Ma, of which, at least 2.5 km of uplift had taken place since 13.8 Ma. These authors based their calculations on the age of youngest sample, the current high physiography of the escarpment and heat flow measurements of Gettings et al. (1986). Therefore, this uplift looks sufficient to reset (U-Th-Sm)/He ages in apatites having closure temperature of ≈70 °C (Braun and van der Beek, 2004) and requiring ~2.5 km of exhumation if surface temperature and geothermal gradient are assumed to be 20 °C and 20 °C/km respectively. However we see that only a smaller number of low elevation coastal samples have reset AFT and AHe ages (Figs. 2.9E, 2.6A-F) which too are primarily from northern parts of our study area (Lines 1 and 2) whereas the majority of samples have old unreset AFT and AHe ages including low(er) elevation samples from southern parts (Lines 3 and 4). This trend is in contrast to the current physiography of the escarpment which has peaks as high as 2750-3200 m in the south and 2100-2400 m in the north (Bohannon et al., 1989). South of our study area, a minimum of 3.5 km of crustal thinning due to vertical displacement across the basal detachment fault is reported from the foothills of Arabian escarpment (Bohannon, 1986). Thus it increases the likelihood that in spite of having higher physiography, the southern coastal samples (Lines 3 and 4) might have observed lesser exhumation due to their downward movement along normal faults. Voggenreiter and Hötzl (1989) also attributed the escarpment uplift to the movement along normal faults in their study of evolution of the same region (southwestern Arabian continental margin), south of our study area. Conversely, the northern samples (Lines 1 and 2), with lower physiography shows relatively higher exhumation as displayed by reset AFT and AHe ages. We attribute this exhumation to erosion as leveling of a landscape by erosion requires exhumation from depths of about five times as the original height e.g. planation of 1000 m of landscape would require exhumation of rocks from ~5km depth (Molnar and England, 1990).

50

Turab, S.A., 2016

Figure 2.9: (A-D) AFT ages from each transect line plotted separately against the distance from the escarpment to observe any along-strike change in escarpment evolution. (E) AFT and average AHe ages of all the samples are plotted against the distance from the escarpment to evaluate the location of youngest AHe age(s) for further interpretation. Escarpment location is at ‘0’ on the horizontal axis and a negative value means distance of the sample location towards the inland plateau.

51

Turab, S.A., 2016

Bohannon et al. (1989) carried out AFT dating along the southwestern Arabian Shield and escarpment from Jeddah to Jizan. Majority (35 out of 44) of their samples gave pre-rifting AFT ages except few which too are roughly concentrated in two areas, one in the southeast between Abha and Jizan and the second in northwest between Maysaan and Al-Lith. Reset AFT ages from our data set, when plotted on the map together with Bohannon et al., (1989) reset samples (< ~32 Ma) and results of stream power analysis, show a good match of reset

AFT ages with streams having higher ‘ksn’ values in both these areas (yellow dashed polygons in Fig. 2.8). However, the coastal plain and escarpment segment between these two regions of high ksn and reset AFT ages shows lower ksn and older AFT age values. Hence it shows that there has been a good match between the information provided by stream power analysis and AFT data and that there has not been enough (uniform) uplift to completely reset the AFT ages throughout the Arabian escarpment.

AFT ages generally display positive correlation towards elevation with exceptions of two low elevation samples (39 and 40) showing old ages (Fig. 2.6A). These old ages are probably the consequence of presence and movement along rift related normal faults. All the samples collected from elevations of >200m show a pre-rifting AFT age (>~32 Ma) suggesting that at the time of initiation of rifting and subsequent uplifting, all these samples were lying either in or above apatite fission track partial annealing zone (PAZ). Hence these samples may show ‘mixed ages’. However, these old ages may record older events of uplift and erosion resulting from intraplate deformation due to collision of Gondwana with Laurussia during Devonian- Early Carboniferous (Bojar et al., 2002) and epeirogenic movements that took place in pre- mid-Tertiary time throughout the northeastern Afro-Arabian continent (Sahagian, 1988). Based on lithostratigraphic data, d’Almeida (2010) reported that subsidence of Red Sea rift, and by inference the uplift of its margins, started in the Cretaceous and probably as early as in the Jurassic. This could very well explain the ages ranging from 45.6 – 177.4 Ma (Table 2.2). However, being unrelated to the Red Sea rifting and subsequent uplifting, these >~32 Ma ages are not discussed in details here. Apart from the aforementioned two samples (39 and 40), samples with elevations of <200m all show post rifting AFT ages (Figs. 2.10, 2.6A; Tables 2.1, 2.2), showing that these samples are brought from below PAZ as a result of exhumation along the Red Sea rift flank thus implementing that these show ‘cooling ages’. Therefore we interpret that current elevation of ~200 m may represent bottom of the Paleo-PAZ at the start of rift related uplift. The amount of uplift could be calculated by translating the thermal information to depth below the surface. This could be achieved by either knowing or assuming geothermal gradient for the area. We here assume 20 °C/km as average value for 52

Turab, S.A., 2016 paleo-geothermal gradient for our samples which is primarily taken from the heat flow measurements over the coastal plain of western Saudi Arabia (Gettings et al., 1986), calculating to the value of 18 °C/km for Arabian shield rocks (Bohannon et al., 1989). Our assumed value is also in good agreement with heat flow measurements in the Red Sea (Makris et al., 1991a) and the value of geothermal gradient (20 °C/km) used for Gulf of Suez and northern Red Sea (Steckler and Omar, 1994). At this gradient the annealing temperature (≈110 °C; Braun and van der Beek, 2004) correspond to depth of ~4.5 km, assuming a 20 °C surface temperature. Thus our reset samples currently at elevations of <~200m show an exhumation of about 4.5 km since the start of rift related uplift. This amount of exhumation shows good agreement with results from previous studies conducted at Red Sea continental margins (Kohn and Eyal, 1981; Bohannon et al., 1989; Omar et al., 1989; Steckler and Omar, 1994; Davison et al., 1994; Bojar et al., 2002; Balestrieri et al., 2005). However, on local scale, early rift may have possessed higher geothermal gradients with hot circulating fluids under volcanic cover (Schmidt et al., 1983). Thus amount of exhumation would be small and difficult to determine as the young AFT ages would basically record the tectonic denudation of volcanic lid and associated removal of hot fluids.

Figure 2.10: Reset AFT and average AHe ages from this study plotted along with the reset AFT ages of Bohannon et al. (1989). For clarity on samples with <500 m elevation and < 50 Ma ages are shown here. Reference to this figure is provided on figure 2.6A as green rectangle.

53

Turab, S.A., 2016

Dating the initiation of uplift is not easy but maximum age for most of the uplift of a series of samples is reported by the youngest AFT age (Bohannon et al., 1989). Thus the youngest AFT age of our data set i.e. 13.2 Ma marks the upper limit for most of the uplifting which may have started earlier. This interpretation is also supported by the fact that all coarse grained clastic rocks, indicative of presence of high relief marginal to the Red Sea, are much younger than rifting. Evidence of high relief in the Red sea marginal areas comes from Pliocene Gasus Formation (Issawi et al., 1971) in eastern Egypt and poorly dated alluvial cobble and boulder conglomerate of Bathan Formation in central Red sea coastal region. The Bathan Formation lies unconformably on top of the early Miocene (Madden et al., 1983) Baid Formation that is associated with early stage of rifting. Thick continental beds encountered in drill holes provide additional evidence for this claim as they overlie the evaporite deposits and middle Miocene Globigerina-bearing strata everywhere. Within the limit of analytical uncertainty, our youngest AFT age of 13.2 ± 1.7 Ma is similar to the youngest AFT age (i.e. 13.8 Ma) reported by Bohannon et al. (1989) for the same area. Using AHe ages from conjugate Eritrean Red Sea margin and predictions from forward modelling, Balestrieri et al. (2005) stated that main phase of post-break up erosion started at around 15 Ma. Collision of Arabian plate with Eurasia in early Miocene (Mouthereau et al., 2012) and subsequent Zagros overthrusting is also held responsible as one of the reasons for the uplift of rocks along southwestern Arabian plate bordering Red Sea (Bohannon et al., 1989; Stern and Johnson, 2010). Therefore the majority of uplift happening post middle Miocene might be due to crustal thickening at the northeastern edge of Arabian plate which increased an ongoing but slower uplift in southwest at Red Sea margin. From the relationship between dispersed and strongly correlated AFT ages on age vs elevation plot, Bohannon et al. (1989) inferred the initiation of uplift at about 20 Ma. These authors also divided the samples from the area between escarpment and the coast into two ‘fields’ based on the fact that uniform cooling should produce simple linear distribution of ages with respect to elevation. Although our data match well with those two ‘fields’ identified by Bohannon et al (1989) and show more or less similar age-elevation trend (Fig. 2.6D) but we do not go on to classify our samples as belonging to two groups because it is unrelated to the principal objective of this paper and also our samples are smaller in number to be confidently classified.

Track lengths from four of our samples are given in figure 2.7. Mean Track lengths (MTL) show a shorter range of values (<12µm) indicating reasonable amount of track annealing for these samples. Samples 31 and 34, from elevations of >2200m with AFT ages of 352 and 331 Ma respectively, show similar MTL (11.3 µm) indicating that these samples were residing 54

Turab, S.A., 2016 well above the apatite PAZ at the time of Red Sea rifting and subsequent uplift. However at the same time sample 37, with AFT age of 72.3 Ma and elevation of 1598 m, and sample 46, with AFT age of 120.5 Ma and elevation of 875 m, were most probably residing with in the apatite PAZ. Track length distribution of these samples also support this fact with lower MTL (10.6 µm) for sample 37 and bimodal distribution for sample 46 with older tracks experiencing substantial annealing and track shortening and new tracks showing longer track lengths (MTL = 11.6 µm).

In summary, our data might sense late cretaceous epeirogenic (70 – 100 Ma AFT ages of our samples) and older (late Devonian to early carboniferous / 331 and 352 Ma) uplift. From at least late cretaceous to onset of rifting, the entire northeastern part of Afro-Arabian continent remained near sea level (Bohannon et al., 1989). Our data suggests ~4.5 km of uplift in the southern Red Sea escarpment region of Saudi Arabia that probably began at about 20 Ma (Bohannon et al., 1989) and accelerated since 13.2 Ma (this study). Sinai (Kohn and Eyal, 1981), eastern dessert of Egypt (Bojar et al., 2002; Omar et al., 1987), southern Red Sea margin of Yemen (Menzies et al., 1997; Davison et al., 1994) and the conjugate margin of southern Red Sea in Eritrea (Balestrieri et al., 2005; Abbate et al., 2002; Ghebreab et al., 2002) also show similar results indicating cooling of Red Sea margins in Neogene possibly beginning in Oligocene to early Miocene.

Escarpment Evolution

Forward modeling predicts AFT (Gallagher et al., 1998) and AHe (Persano et al., 2002; Braun and van der Beek, 2004) ages along continental margins (Figs. 2.1B,C). Their distribution from coast to inland plateau across the coastal plain and along the escarpment can thus be used to constraint the escarpment evolution provided there has been at least 2 km of post break-up uplift and erosion. Constraints on the timing of escarpment formation at its present location and geometry are provided by the minimum AHe observed on the coastal plain (Braun and van der Beek, 2004). Thus minimum rate of escarpment evolution could be determined by analyzing samples along a transect from the base of the escarpment to the coast.

The downwarp model predicts both AFT and AHe ages at the coast that are older than the age of break up and becomes younger towards the escarpment due to progressive increase in amount of denudation (Fig. 2.1). The escarpment retreat scenario forecasts the minimum age 55

Turab, S.A., 2016 near the present-day escarpment, whereas in the case of plateau degradation, the minimum age is found halfway between the coast and the escarpment due to isostatically driven exhumation, which is always intense at the center of the region of erosional unloading (Braun and van der Beek, 2004).

AFT ages when plotted against the distance of samples from the escarpment-lip displays negative correlation displaying oldest ages at the top of the escarpment while the young, reset, ages lie at distance (33-92 km) from the escarpment on the coastal plains with samples 39 and 40 lying as outliers (Fig. 2.9E). Along strike variation of AFT vs distance from the escarpment could also be seen when AFT age data from individual profile lines is plotted against distance from the escarpment separately (Fig. 2.9). AFT ages of Line 1 samples decrease from 138 Ma in inland plateau to 96 Ma at the escarpment-lip to 17 Ma near the Red Sea coast about 92 km from escarpment-lip (Fig. 2.9A). Line 2 and Line 3 samples also display a similar trend of AFT age decrease from escarpment-lip towards the coast (Figs. 2.9B,C). However, Line 4 samples show a decreasing trend in the AFT ages from the escarpment towards the coast until about half way whereas the two samples close(r) to the coast show old (275 and 280 Ma / Permian) ages (Fig. 2.9D). This may well be due to normal faulting reported from southern Red Sea area (Bohannon, 1986) or due to the localization of hinge between continually subsiding shelves and uplifting continents which reportedly lies in the coastal plains along the southern Red Sea and near the coastline in the north (Bohannon et al., 1989). Occurrence of old tilted flows (25-30 Ma) at low elevations is also used as an evidence for rift related normal faulting (Bohannon et al., 1989). Thus the distribution of AFT ages (Fig. 2.9E) allows us to reject the downwarp model of escarpment evolution (Fig. 2.1B). However, as the differences among the patterns of AFT ages of downwearing and scarp retreat models across the coastal plain are not large (Fig. 2.1B), we use the distribution of AHe ages to help constrain the escarpment evolution model. Contrary to the similar AHe age distribution throughout the coastal plain predicted by downwearing scenario, the Escarpment retreat scenario predicts a well-defined trend of decreasing AHe ages from coast towards the escarpment base (Fig. 2.1C) implementing that the youngest AHe ages would be located near the base of current escarpment. Detailed examination of our AHe age data set shows that it does not follow a younging trend from coast to escarpment base and also the location of our youngest AHe ages is about 44-52 km from escarpment, roughly at the center between coast and escarpment (Fig. 2.9E). We thus interpret that these facts are supporting plateau degradation or downwearing model of escarpment evolution where exhumation is greatest half way between coast and escarpment (Braun and van der Beek, 2004). 56

Turab, S.A., 2016

Constraints could be placed on the escarpment establishment at its present location by using the ages of lava flows that erupted in the elevated plateau, flowed in the paleo valleys of the escarpment and covered coastal areas at low elevations indicating that escarpment was already in place before that time. Such flows exist in the northeast of Jeddah at Harrat Rahat basalt field extending from elevations of about 1250 m to lower than 200 m on the coastal plain. Though most of these flows are younger than 4.5 Ma, but important for our consideration are the oldest reported dates of 12.6 and 13.2 Ma (Brown et al., 1989; Coleman et al., 1983). Pallister (1987) also reported a similar flow from the southeast of Jeddah (near Jabal Sita) that once filled a paleovalley within the Red Sea trough and yields a whole rock K- Ar age of 7 Ma and a clinopyroxene age of 11.3 Ma. Thus we can argue that escarpment was in place at its present day location at around 10 Ma.

Concept of Rifting/Active Vs Passive Rifting

Uplifting has been dominant on the Arabian side than the Nubian side (in Africa) as evidenced by physiographic asymmetry across Red Sea with average elevations in former being slightly higher than latter (Fig. 2.11). Both Arabian and African margins of Red Sea show a general decrease in average elevations from south towards north (Bohannon et al., 1989). Voggenreiter and Hötzl (1989) used the distribution of Tertiary and Quaternary volcanics which are almost exclusively confined to the Arabian Peninsula, to highlight asymmetry and argued that Red Sea rifting could have taken place along low angle, normal shear zone by simple-shear mechanism. Collision of Arabian plate with Eurasia and subsequent Zagros overthrusting resulting in the crustal thickening may also explain the uplift of southwestern margin of the Arabian plate bordering Red Sea (Bohannon et al., 1989; Stern and Johnson, 2010). Thus, the resultant tilt at the base of the Arabian plate would have further facilitated the flow of basal fluid towards the rift and would result in partial melting of mantle lithosphere enhancing the uplift process (Fig. 2.11).

There have been different rifting models put forward for Red Sea rifting e.g. passive continental rifting (Omar et al., 1989; Bohannon et al., 1989), active continental rifting (Menzies et al., 1997) and those differing from both active and passive models (Bosworth et al., 2005; Bojar et al., 2002; Omar and Steckler, 1995; Davison et al., 1994). However we here compare our results against active or passive rift models initially introduced by Sengör and Burke (1978). These models follow a sequential development of rift margins with

57

Turab, S.A., 2016 differences in the sequence of events i.e. the active rifting follows a doming(uplift)- volcanism-rifting sequence as it is driven by plume whereas passive rifting takes place in rifting-uplift-volcanism sequence.

Figure 2.11: Schematic cross section across the Red Sea rift basin showing the present-day configuration. Tilt at the base of the Arabian plate results from crustal thickening due to Zagros overthrusting. This leads to partial melting of the lithospheric mantle under the Arabian escarpment region as hot melts are moving towards the rift axis, thus resulting in the asymmetric uplift of the Arabian margin as compared to the conjugate African margin. Topography of the escarpment regions (bold lines) are calculated by myself as part of swath profiles (Fig. 2.3); Topography/bathymetry for the Red Sea and location of Moho under the Red Sea and escarpment region of Ethiopia is taken from Makris et al. (1991b); Topography for central Arabia is taken from Gettings et al. (1986); Location of Moho under Arabian escarpment and central Arabia is from Mooney et al. (1985). Moho under Ethiopian side is hypothetically extended based on information from Chang and Van der Lee (2011). Topography towards the inland side of African escarpment and base of the lithosphere is shown hypothetically.

In southern Red Sea, the oldest rift related volcanic rocks (about 29-31 Ma) are strongly deformed and tilted but are intruded by 20-23 Ma gabbro and granophyre (Coleman et al., 1977) and slightly older dikes that show very little sign of deformation (Bohannon et al., 1989). Similarly, 32-25 Ma, oldest basalt flows near Jeddah are strongly deformed whereas 21 Ma and younger dikes, flows and plutons are undeformed. There was no significant regional extension associated with early magmatism and an extreme phase of extension might have taken place around 25 Ma (Bosworth et al., 2005) with the overall normal and detachment faulting activity taking place at 23-29 Ma (Bohannon, 1986) whereas there has been no record of prevolcanic rifting (Bohannon et al., 1989).

Baid, the oldest rift related formation between Jeddah and Jizan, is fine grained and rarely thicker than 300 m. Lacustrine depositional and low relief source environment with soil cover along the rift margins is indicated by the lithology of Baid formation. It is tilted due to normal faulting and is intruded by 18.4-19.7 Ma basalt sills. Baid formation is interbedded with and considered laterally equivalent to the volcanic rocks that are older than 20 Ma (Schmidt et al., 1983). In Gulf of Suez, upper part of the oldest syn-rift sedimentary unit (Nukhul formation) is ca. 19-23 Ma while the base, being poorly dated, may extend into the Oligocene (Omar et al., 1989). Though emergent but there has been no geological evidence(s) to support any large 58

Turab, S.A., 2016 uplifts in central and southern parts of Arabia during early phase of rifting e.g. all of the marine sedimentary rocks of Eocene and early Oligocene age from the surroundings are fine grained, so the difference in elevation between the areas of marine deposition and of soil development was small. Geological evidences suggest that the entire northeastern part of Afro-Arabian continent was near sea level (Gulf of Suez) or a low relief zone of thick laterite soils (southern Red Sea) for 45 Myr prior to the onset of Red Sea rifting thus neglecting any pre-rift uplift or doming (Bohannon et al., 1989; Steckler and Omar, 1994; Omar et al., 1989).

The oldest angular unconformities occur beneath 15-18 Ma flows implementing that a measurable extension had occurred by the middle Miocene. There was very little uplift of Tertiary age prior to that time as evident from rift-related fine grained marine sedimentary rocks. Thus Our data, along with those of Bohannon et al. (1989) from southwestern Saudi Arabia, Kohn and Eyal (1981) from Sinai and Omar et al. (1987) from Egypt, support that post middle Miocene uplift around the Red Sea is sufficient to account for all of the present- day high elevations.

Small amounts of alkaline volcanism began 30-32 Ma that has increased in volume till present. It is difficult to date the initiation of extension but it did not start prior to volcanism and was over by 20-24 Ma (Bohannon et al., 1989; Bosworth et al., 2005). Whereas uplift is clearly constrained by youngest AFT age of 13.2 Ma and abundance of AFT data around Red Sea margins to be related to later stages (post middle Miocene) of geological evolution of the area. As active mantle hypothesis could be rejected based on the absence of pre-rift doming/uplift, it is also difficult to argue in support of passive mantle hypothesis because the geologic history described above does not follow the suggested sequence of events i.e. rifting- uplift-volcanism. The sequence of events was magmatism most probably accompanied by synchronous extension and followed by uplifting which do not fit either active or passive rifting models (e.g. Bosworth et al., 2005; Bojar et al., 2002; Omar and Steckler, 1995; Davison et al., 1994). In summary, the Red Sea rifting seems to have resulted from a complex interaction between active and passive rift processes, as proposed by Courtillot et al. (1999) and Hill (1991).

59

Turab, S.A., 2016

CONCLUSIONS

Following conclusions can be made from the results of low temperature thermochronological study at the Red Sea escarpment region of southwestern Saudi Arabia:

Contrary to the present-day physiography, the amount of exhumation in the studied area increases from south towards north. Exhumation is controlled by erosion in the north (line 1 and 2) and probably by tectonics (normal faults) in the south as evidenced by two samples (39 and 40) having old AFT ages.

The bottom of paleo-PAZ is interpreted to be at ~200m present day elevation as all the samples collected from elevations of >200m show a pre-rifting AFT age (>~32 Ma) suggesting that at the time of initiation of rifting and subsequent uplifting, all these samples were lying either in or above apatite fission track partial annealing zone (PAZ).

There has not been enough (uniform) uplift throughout the study area to completely reset the AFT and AHe ages of all the samples. Rather, the higher amount of exhumation is confined to two separate regions separated by a region with lower amount of exhumation. This interpretation is also supported by stream power analysis of the region. However, the reset AFT ages indicate about 4.5 km of exhumation which although may have started earlier (early Miocene) but most of it took place after 13.2 Ma, the youngest age of our data set.

Distribution of AFT and AHe ages across the Red Sea escarpment and coastal region of southwestern Saudi Arabia shows that the escarpment evolved by the downwearing of an elevated plateau between present day escarpment and continental margin towards the coast.

Geological evidences from around the entire Red Sea area suggest that Red Sea rifting did not follow the suggested sequences of events by either active or passive rifting models. Rather, Red Sea rifting may have resulted from the complex interaction of both active and passive rifting models.

60

Turab, S.A., 2016

Chapter 3.

Overall Conclusions

61

Turab, S.A., 2016

In this thesis I present the results of a study carried out to constrain the amount and understand the processes of rock exhumation in two examples of mountain ranges from two contrasting tectonic environments: The western Himalayan syntaxis region as an example for a compressional mountain belt and the Great Escarpment of Saudi Arabia as an example for a mountain belt formed in an extensional environment. As the focus of this thesis is on the exhumation of rocks through the last few kilometers of crust, I have used a combination of several low temperature thermochronologic techniques such as U-Pb, fission track and (U-Th- [Sm])/He on apatite and (U-Th)/He on zircon along with the geomorphic analysis to quantify the rock exhumation through the upper crust.

In northwest Himalayas, apatite U-Pb data allow us to constrain the position of the MCT which has been the subject of controversy. The two apatite low-temperature thermochronology systems are largely indistinguishable from each other as average ZHe, AFT and average AHe ages are generally 9 – 16 Ma, 3 - 7 Ma and 3 – 6 Ma, respectively. They indicate exhumation rates of 1 mm/year to 0.3 mm/year. Thermal history modelling confirms the MBT thrusting around 10 Ma in both Kel and Leswa sections and also that the MCT has remained inactive since at-least 11-10 Ma. Stream power analysis shows strong geomorphic disequilibrium caused by recent activity along the MBT/MCT and continuous catchment capture in the headwaters. Knick points have been identified along the channel profiles of selected rivers which indicates a wave of erosion propagating upstream that may have started as a result of base-level fall at 5-3 Ma. In combination with apatite fission track ages from the literature, our data allow us to conclude that the exhumation age contours trend parallel to regional-scale thrust faults rather than forming an isolated “tectonic aneurysm” as has been suggested in recent literature. This implies that exhumation is likely to be controlled by the crustal-scale thrust rather than fluvial erosion.

In the Red Sea escarpment region of southwestern Saudi Arabia, the amount of exhumation increases from south towards north and is controlled by erosion in the north and probably by tectonics in the south. Bottom of the paleo-PAZ present at the time of initiation of rifting and subsequent uplifting is interpreted to be at ~200 m present day elevation. AFT and AHe ages of all the coastal plain samples are not reset indicating that there has not been enough (uniform) uplift throughout the study area. Rather, the higher amount of exhumation is confined to two separate regions separated by a region with lower amount of exhumation. This interpretation is also supported by stream power analysis of the region. However, the reset AFT ages indicate about 4.5 km of exhumation which although may have started earlier

62

Turab, S.A., 2016

(early Miocene) but most of it took place after 13.2 Ma, the youngest age of our data set. The distribution of AFT and AHe ages shows that the escarpment evolved by the downwearing of an elevated plateau between present day escarpment and continental margin towards the coast. Geological evidences suggest that Red Sea rifting did not follow the suggested sequences of events by either active or passive rifting models. Rather, Red Sea rifting may have resulted from the complex interaction of both active and passive rifting models.

In summary, the results of this study indicate that the exhumation of rocks in extensional and compressional mountain belts is controlled by dissimilar processes. Tectonics plays the major role in exhuming rocks in northwest Himalayas whereas at the Red Sea escarpment of southwestern Saudi Arabia, erosion seems to be controlling the exhumation of rocks after initial rifting and subsequent rift flank uplift. Thus the results of this study may also be applicable generally to other examples of extensional and compressional mountain belts.

63

Turab, S.A., 2016

Appendix A

64

Turab, S.A., 2016

This appendix contains all the abstracts (either published or submitted) at various international conferences. The list begins with the oldest and ends with most recent work.

An integrated approach to study the exhumation of rocks in Neelum valley, NW Himalayas, Pakistan.

Turab, S.A., Stüwe, K., Stuart, F.M., and Chew, D.M., 2016, Geophysical Research Abstracts, v. 18, EGU2016-4788, EGU General Assembly, 17-22 April, Vienna, Austria.

Tectonics and erosion have both been suggested as alternative driving mechanisms for rapid exhumation of rocks in the western Himalayan syntaxis. This debate could be resolved by understanding the plan view-geometry of the exhumation of rocks in the region: does it follow the major structures?, or is it related to the drainage geometry? In order to resolve this geometry we have undertaken a low-temperature thermochronologic study, using crystalline rocks, of a critical region of the western syntaxis: Neelum valley region, Pakistan. Apatite (U- Th-Sm)/He (AHe), fission track (AFT) and U-Pb dating has been combined with geomorphic stream power analysis in order to discern the relationship of exhumation of rocks to tectonics (main faults) or erosion. Pooled AFT ages show a range of 2.2 ± 0.4 to 7.0 ± 0.4 Ma (1σ). Recoil corrected AHe ages exhibit a range from 2.0 ± 0.1 to 8.7 ± 0.5 Ma (1σ). U-Pb ages could be used to divide the samples in three groups: ages that are completely-, partly- and not- affected by Himalayan tectonics. The range of apatite U-Pb ages displayed by both completely- and partly- affected samples is from 17.0 to 43.0 Ma (2σ, unanchored, i.e. constrained by isochrones alone) and 6.0 to 48.3 Ma (2σ, anchored using the Stacey and Kramers terrestrial Pb evolution model). Stream power analysis of the Neelum river catchment indicates a region with high steepness index (Ksn, normalized to reference 0.9 concavity, θref = 0.45) values of > 500 m which coincides well the region sampled.

In combination with earlier published ages, our data indicate that exhumation contours run more or less parallel to the major structures in the region. The boundary between samples with unaffected and affected U-Pb ages as well as transition from high Ksn to lower Ksn values along the main Neelum river fits well with the mapped trace of the Main Central Thrust (MCT), corroborating the presence of the MCT in the southeastern parts of our study area. Thermal history modeling of the AFT and AHe ages indicates recent rapid exhumation of rocks through the upper 5 - 6 km of the crust. Also, the AFT ages are younger in proximity to the main faults. This new data supports a model with tectonics as the main driving mechanism for exhumation of rocks and may indicate quite recent re-activation of MCT or one of the other major structures near it.

65

Turab, S.A., 2016

Constraining Uplift and Erosion in the Western Himalayan Syntaxis. A Multiple Thermochronometric Study from the Neelum River Region, NW Himalayas, Pakistan

Turab, S.A., Stüwe, K., Stuart, F.M., Chew, D.M., and Cogné, N., 2016, 15th International Conference on Thermochronology (Thermo-2016), 18-23 September, Maresias, Brazil.

Tectonics and erosion have been suggested as alternative driving mechanisms for rapid exhumation of the western Himalayan syntaxis. The geometry of the region showing rapid exhumation holds this key information. Main agent of rapid exhumation could thus be nominated by understanding whether the exhumation follows the major structures (tectonics) or the drainage pattern (erosion) (Fig. 1A). In order to resolve this debate we combine (U-Th- Sm)/He, fission track and U-Pb dating of apatite with (U-Th)/He dating of zircon and stream power analysis from the Neelum valley region of Azad Jammu and Kashmir, Pakistan (Fig. 1B). Pooled fission track ages show a range of 2.2 ± 0.4 to 7.0 ± 0.4 Ma (1σ) while apatite He ages range from 2.0 ± 0.1 to 8.7 ± 0.5 Ma. Zircon He ages show a range from 6.1 ± 0.1 to 14.6 ± 0.3 Ma and apatite U-Pb ages are Proterozoic to as young as 17.0 Ma and can be separated into three groups depending on the degree they have been affected by Himalayan tectonics. Stream power analysis of the Neelum river catchment indicates high steepness index of Ksn > 500 m0.9 along the major river and lower Ksn in the headwaters. The boundary between samples with unaffected and affected U-Pb ages, as well as transition from high Ksn to lower Ksn values along the main Neelum river, fit well with the mapped trace of the Main Central Thrust (which has been a topic of controversy), thus corroborating the presence of the thrust in the southeastern parts of our study area. Thermal history modeling indicates accelerated exhumation at ~10 Ma in the uppermost 5-6 km of the crust that might represent the onset of movement along Main Boundary Thrust. In combination with published cooling ages, our data indicate that exhumation contours run more or less parallel to the major structures in the region and fission track ages are younger close to the main faults supporting the contention that tectonics is the main driving mechanism for exhumation at the western syntaxis of the Himalayas (Fig. 1C).

66

Turab, S.A., 2016

Figure 1. Geological map (A) showing predictions for areas with high exhumation rates. Red text highlights different rock units (after Wilke et al., 2012). Yellow ellipses show the location of two profiles sampled; (B) map of the same area as (A), showing regional scale faults and rivers along with published apatite fission track ages. Green dots and contours = after Zeitler (1985) also showing region of high exhumation rates (green area). Red dots = after Wilke et al. (2012) and references therein. Purple = van der Beek et al. (2009). Yellow transparent ellipse highlights a region without data and hence tackled herein this study; (C) Same map as (B) with the addition of new apatite fission track data from this study and re-interpretation of exhumation age-contours.

References 1. Van der Beek, P., Van Melle, J., Guillot, S., Pêcher, A., Reiners, P. W., Nicolescu, S. & Latif, M. Eocene Tibetan plateau remnants preserved in the northwest Himalaya. Nature Geoscience 2, 364-368 (2009). 2. Wilke, F. D. H., Sobel, E. R., O’Brien, P. J. & Stockli, D. F. Apatite fission track and (U–Th)/He ages from the Higher Himalayan Crystallines, Kaghan Valley Pakistan: Implications for an Eocene Plateau and Oligocene to Pliocene exhumation. Journal of Asian Earth Sciences 59, 14-23 (2012). 3. Zeitler, P. K. Cooling history of the NW Himalaya, Pakistan. Tectonics 4 (1), 127-151 (1985). 67

Turab, S.A., 2016

Escarpment evolution at the Red Sea continental margin of southwestern Saudi Arabia

Turab, S.A., Stüwe, K., Stuart, F.M., Chew, D.M., and Cogné, N., 2017, submitted for EGU General Assembly, 23-28 April, Vienna, Austria.

Rifting of the Red Sea started around ~30-32 Ma and resulted in the formation of one of the youngest and best developed escarpments of the world: the Great Escarpment of southwestern Saudi Arabia. The escarpment is perfectly developed over a length of more than 500 km and includes mountains up to 3000 m in elevation. To better understand the geodynamics of Red Sea rifting and to constrain a denudational model for the Great Escarpment, the results of fission track and (U-Th-[Sm])/He thermochronologic techniques on apatite are combined with stream power analysis of the central part of this region. Pooled fission track ages (recording cooling through about 110 °C) range from 13.2 ± 1.7 to 352.1 ± 17.6 Ma (1σ) with all ages that are younger than about 50 Ma (and thus related to the rifting) being from elevations lower than about 500 m at the base of the escarpment. Apatite He ages range from 2.8 ± 0.3 to 264.5 ± 19.6 Ma with a similar age-elevation relationship. Bottom of the pre uplift partial annealing zone is interpreted to be lying at ~200 m present-day elevation. Our fission track data indicate that the amount of exhumation is insufficient to completely reset all the coastal plain samples, but exhumation along the escarpment appears to increase from south towards north. Highest amount of exhumation is confined to two separate regions, one in the north and second in the south, which are separated by a region of non-reset AFT ages and hence lower amount of exhumation. This interpretation is also supported by stream power analysis of the region. Reset AFT ages indicate about 4.5 km of exhumation which may have started in early Miocene but the majority of which occurred after 13.2 Ma. This interpretation is consistent with a single isolated outcrop of Nubian sandstone at the summit of Saudi Arabia’s highest peak. Distributions of AFT and AHe ages across the escarpment and coastal region supports the escarpment development by the established “downwearing” or “plateau degradation” model of escarpment evolution, which implies that the present drainage divide may have been in its position already in the Miocene. Red Sea rifting may have resulted from the complex interaction of both active and passive rifting models as it did not follow the suggested sequence of events by either of the two rifting models.

68

Turab, S.A., 2016

Appendix B

69

Turab, S.A., 2016

This appendix contains details about the standard lab procedures and some relevant figures which are also submitted as supplementary material along with chapter 1. Details about all the samples collected as part of this PhD study is also provided in the form of two tables.

Standard Lab Procedures

Apatite and zircon were separated from rock samples by crushing, sieving, gravity separation using Wilfley table, magnetic separation and heavy liquid separation using sodium polytungstate and di-iodomethane at University of Graz. AFT analyses were carried out at

Trinity College Dublin. Samples were etched in 5.5 M HNO3 for 20 seconds at 21°C (after Donelick et al., 2005). Laser-ablation inductively coupled plasma mass spectrometry (LA- ICP-MS) was used to determine uranium concentrations for fission-track dating (Donelick et al., 2005; Chew and Donelick, 2012). These analyses also provided apatite Cl concentrations (Chew et al., 2014a) and apatite U–Pb age data (Fig. B1; Chew et al., 2014b).

For apatite (AHe) and zircon (ZHe) (U-Th-Sm)/He analysis, single, euhedral, inclusion-free crystals were handpicked, measured and packed in Pt foil tubes. Helium was extracted by heating the Pt foils with a 808 nm diode laser at 600 – 700°C for 60 seconds (apatite) and at ~1200°C for 20 minutes (zircon) (Foeken et al., 2006). Helium was measured using a Hiden HAL3F quadrupole mass spectrometer. Apatite-bearing packets were then removed from the He extraction line, spiked with 235U and 230Th in 5% nitric acid and left at 80°C for 48 hours in sealed Teflon beakers. Zircon crystals were removed from the tubes before being dissolved in 49% HF at 235°C for 48 hours in a Parr bomb (Dobson et al., 2008). 238U, 235U and 232Th contents were determined via isotope dilution ICP-MS (Balestrieri et al., 2005). Durango apatite and Fish Canyon Tuff zircon were used as mineral age standards. The AHe and ZHe ages were calculated according to established procedures (Meesters and Dunai, 2005), and the ages corrected for alpha recoil (Farley et al., 1996; Ketcham et al., 2011).

70

Turab, S.A., 2016

Figure B1: Tera-Wasserburg Concordia diagrams for all the samples dated using U-Pb method on apatite. Note that for partly reset samples (i.e. N5, N7, N8 and L2) ellipses with different colors indicate data from young and old apatite grains whereby data from only young grains is used for age calculation.

71

Turab, S.A., 2016

Figure B2: Thermal history modelling results for Kel and Leswa profiles showing observed versus predicted ages. LL: Log Likelihood.

72

Turab, S.A., 2016

TABLE A1: DETAILS OF ALL SAMPLES COLLECTED FROM NORTHWEST HIMALAYAN SYNTAXIS REGION FOR LOW TEMPERATURE THERMOCHRONOLOGIC ANALYSIS Sample Latitude* Longitude* Sample reference Elevation Rock type (°N) (°E) (m) K1 34 51 30.5 74 17 52.4 Kel 4225 Garnet bearing Gneiss, Medium grained K2 34 51 14.5 74 17 46.3 Kel 3916 Garnet bearing Gneiss, Pegmatite K3 34 51 01.9 74 17 47.1 Kel 3768 Pegmatite K4 34 50 49.2 74 17 51.6 Kel 3492 Garnet bearing Gneiss K5 34 50 35.5 74 17 56.8 Kel 3210 Mica Gneiss K6 34 50 23.7 74 18 03.7 Kel 3040 Pegmatite, Garnet bearing K7 34 50 01.1 74 18 20.5 Kel 2790 Pegmatite K8 34 49 55.4 74 18 32.4 Kel 2476 Pegmatite K9 34 49 36.4 74 18 38.5 Kel 2228 Mica Gneiss K10 34 49 12.9 74 18 22.5 Kel / Neelum 2015 Mica Gneiss, Garnet present N2 34 49 18.1 74 13 12.0 Neelum 1893 Gneiss with large Garnet crystals N3 34 43 17.4 74 03 06.5 Neelum 1729 Garnet bearing Gneiss N4 34 42 46.9 73 58 49.1 Neelum 1581 Garnet bearing Gneiss N5 34 38 52.1 73 56 13.7 Neelum 1469 Granite N6 34 32 51.5 73 50 49.2 Neelum 1296 Para-Gneiss, showing contact between 2 different rock types N7 34 32 04.9 73 50 45.1 Neelum 1285 Granite N8 34 27 46.1 73 49 11.7 Neelum / Leswa 1177 Pegmatitic Gneiss L2 34 27 55.2 73 48 36.9 Leswa 1357 Granitic Gneiss L3 34 27 32.4 73 47 18.9 Leswa 1614 Slate L4 34 27 30.6 73 46 29.1 Leswa 1807 Slate L5 34 27 34.3 73 45 34.9 Leswa 2065 Garnet Mica Schist L6 34 27 41.2 73 45 00.7 Leswa 2329 Gneiss (Mica-Bio + Mus) L7 34 27 25.8 73 44 51.1 Leswa 2544 Gneiss L8 34 26 48.6 73 45 31.3 Leswa 2675 Slate L9 34 26 18.2 73 45 04.8 Leswa 2489 Gneiss L10 34 25 43.7 73 44 15.6 Leswa 2293 Gneiss L11 34 25 47.2 73 44 14.4 Leswa 2290 Pegmatite L12 34 26 51.9 73 42 36.5 Leswa 1890 Volcanic L13 34 26 58.6 73 41 23.3 Leswa 1389 Sandstone-miocene L14 34 26 39.8 73 39 31.4 Leswa 1132 Sandstone-miocene *data entered in “Degrees Minutes Seconds” format.

73

Turab, S.A., 2016

TABLE A2: DETAILS OF ALL SAMPLES COLLECTED FROM SOUTHWESTERN SAUDI ARABIA FOR LOW TEMPERATURE THERMOCHRONOLOGIC ANALYSIS Sample Latitude* Longitude* Elevation Rock type (°N) (°E) (m) 1 20 15.584 40 26.407 81 Pegmatite 2 20 38.375 40 33.190 477 Gabro 3 20 34.105 40 28.014 260 Granite 4 20 27.684 40 28.272 167 Granite 5 20 19.959 40 28.046 109 Volcanic (Basalt), from big boulder 6 19 45.488 40 56.209 67 Granite 7 19 50.417 41 00.142 178 Granite, weathered and feldspar mostly altered 8 19 53.616 41 03.227 180 Granite, nice and fresh 9 19 54.974 41 08.598 248 Granite 10 19 56.138 41 17.679 552 Granite dyke + dark colored host rock 11 19 55.489 41 16.313 471 - 12 19 49.535 41 22.593 447 - 13 19 56.843 41 26.881 858 - 14 19 59.800 41 26.369 1273 Magma mingling (granitic + mafic rock) 15 19 59.521 41 25.771 1629 - 16 20 00.580 41 26.608 1975 - 17 20 03.367 41 23.796 2483 Amphibolites, fresh 18 20 11.013 41 15.106 2135 Quartzite (?), fresh, unaltered 19 20 24.956 41 07.468 2033 Granite 20 20 30.797 41 01.729 1859 Aplite + schist 21 20 39.076 40 59.851 1749 Granite 22 20 43.793 40 49.424 2277 Granite 23 20 46.803 40 57.712 1767 Gneisses + small pegmatite 24 20 58.250 41 04.615 1494 Granodiorite-diorite 25 20 49.618 41 10.104 1445 Granite 26 20 24.249 41 18.899 1772 - 27 20 01.878 41 26.980 2211 Granite 28 19 55.378 41 33.700 2080 Granite, tonalites 29 19 50.021 41 51.685 1897 Syenite (?) 30 19 30.351 41 57.093 2138 Tonalite 31 19 06.561 42 07.759 2398 - 32 18 52.124 42 14.228 2460 Diorite 33 18 31.909 42 25.768 2218 Intrusive felsic rock 34 18 13.255 42 31.333 2221 Granite 35 18 16.975 42 22.105 2912 Sandstone + loose sand 36 18 17.915 42 19.875 2225 dark grey rock with some coarser parts 37 18 16.771 42 19.945 1598 Dark greyish rock with some light colored minerals, fallen pieces from outcrop 38 17 53.674 42 15.069 169 Pegmatite/Granite intruding foliated metamorphic rock of staurolite grade 39 17 52.176 42 12.516 116 Aplite, with little or no biotite, from road cut boulders 40 18 07.273 41 38.745 40 Schist + granite 41 19 01.876 41 28.769 121 Needle-like-amphibole rich rock 42 19 05.828 41 33.207 147 Granite, fresh, from boulder 74

Turab, S.A., 2016

43 19 06.269 41 38.100 212 Granite 44 19 04.715 41 47.607 289 Granite 45 18 47.624 41 59.198 393 Granite, tonalite 46 19 07.219 42 03.857 875 Granite, fresh 47 19 05.552 41 57.002 491 Syenite 48 19 24.726 41 44.967 616 Needle like amphibole 49 19 42.773 41 24.036 313 Granite, fresh, from quarry 50 20 25.098 39 58.410 42 Granite, weathered, in desert *data entered in DDM (Degrees Decimal Minutes) format.

75

Turab, S.A., 2016

References

Abbate, E., Balestrieri, M.L., and Bigazzi, G., 2002, Morphostructural development of the Eritrean rift flank (southern Red Sea) inferred from apatite fission track analysis: Journal of Geophysical Research, v. 107, no. B11, 2319, doi:10.1029/2001JB001009.

Alam, A., Ahmad, S., Bhat, M.S., and Ahmad, B., 2015, Tectonic evolution of Kashmir basin in northwest Himalayas: Geomorphology, v. 239, p. 114-126, doi:10.1016/j.geomorph.2015.03.025.

Balestrieri, M.L., Stuart, F.M., Persano, C., Abbate, E., and Bigazzi, G., 2005, Geomorphic development of the escarpment of the Eritrean margin, southern Red Sea from combined apatite fission-track and (U–Th)/He thermochronometry: Earth and Planetary Science Letters, v. 231, p. 97–110, doi:10.1016/j.epsl.2004.12.011.

Blenkinsop, T., and Moore, A., 2013, Tectonic geomorphology of passive margins and continental hinterlands, in Shroder, J., (Editor in Chief), Owen, L.A., (Ed.), Treatise on Geomorphology: Academic Press, San Diego, CA, v. 5, Tectonic Geomorphology, p. 71–92.

Bohannon, R.G., 1986, Tectonic configuration of the western Arabian continental margin, southern Red Sea: Tectonics, v. 5, no. 4, p. 477-499.

Bohannon, R.G., Naeser, C.W., Schmidt, D.L., and Zimmermann, R.A., 1989, The Timing of Uplift, Volcanism, and Rifting Peripheral to the Red Sea: A Case for Passive Rifting?: Journal of Geophysical Research, v. 94, no. B2, p. 1683-1701.

Bojar, A.-V., Fritz, H., Kargl, S., and Unzog, W., 2002, Phanerozoic tectonothermal history of the Arabian–Nubian shield in the Eastern Desert of Egypt: evidence from fission track and paleostress data: Journal of African Earth Sciences, v. 34, p. 191–202.

Bosworth, W., Huchon, P., and McClay, K., 2005, The Red Sea and Gulf of Aden basins: Journal of African Earth Sciences, v. 43, p. 334–378, doi:10.1016/j.jafrearsci.2005.07.020.

Braun, J., and van der Beek, P., 2004, Evolution of passive margin escarpments: What can we learn from low-temperature thermochronology?: Journal of Geophysical Research: Earth Surface, v. 109, F04009, 14 p., doi:10.1029/2004JF000147.

76

Turab, S.A., 2016

Brown, G.F., Schmidt, D.L., and Huffman, A.C., Jr., 1989, Geology of the Arabian Peninsula shield area of western Saudi Arabia: U.S. Geological Survey Professional Paper 560- A, 188 p.

Burbank, D.W., Leland, J., Fielding, E., Anderson, R.S., Brozovic, N., Reid, M.R., and Duncan, C., 1996, Bedrock incision, rock uplift and threshold hillslopes in the northwestern Himalayas: Nature, v. 379, p. 505-510.

Burg, J-P., Chaudhry, M.N., Ghazanfar, M., Anczkiewicz, R., and Spencer, D., 1996, Structural evidence for back sliding of the Kohistan arc in the collisional system of northwest Pakistan: Geology, v. 24, no. 8, p. 739-742, doi:10.1130/0091- 7613(1996)024<0739:SEFBSO>2.3.CO;2.

Butler, R.W.H., Casey, M., Lloyd, G.E., Bond, C.E., McDade, P., Shipton, Z., and Jones, R., 2002, Vertical stretching and crustal thickening at Nanga Parbat, Pakistan Himalaya: A model for distributed continental deformation during mountain building: Tectonics, v. 21, no. 4, 1027, 17 p., doi:10.1029/2001TC901022.

Butler, R.W.H., Prior, D.J., and Knipe, R.J., 1989, Neotectonics of the Nanga Parbat Syntaxis, Pakistan, and crustal stacking in the northwest Himalayas: Earth and Planetary Science Letters, v. 94, p. 329-343.

Calkins, J.A., Offield, T.W., Abdullah, S.K.M., and Tayyab Ali, S., 1975, Geology of the Southern Himalaya in Hazara, Pakistan, and adjacent areas: U.S. Geological Survey Professional Paper 716C, 29 p.

Catlos, E.J., Harrison, T.M., Kohn, M.J., Grove, M., Ryerson, F.J., Manning, C.E., and Upreti, B.N., 2001, Geochronologic and thermobarometric constraints on the evolution of the Main Central Thrust, Central Nepal Himalaya: Journal of Geophysical Research, v. 106, no. B8, p. 16,177–16,204.

Chamberlain, K.R., and Bowring, S.A., 2000, Apatite-feldspar U–Pb thermochronometer: A reliable, mid-range (~450 °C), diffusion-controlled system: Chemical Geology, v. 172, no. 1-2, p. 173-200.

Chang, S.-J., and Van der Lee, S., 2011, Mantle plumes and associated flow beneath Arabia and East Africa: Earth and Planetary Science Letters, v. 302, no. 3-4, p. 448-454, doi:10.1016/j.epsl.2010.12.050.

77

Turab, S.A., 2016

Chemenda, A.I., Mattauer, M., Malavieille, J., and Bokun, A.N., 1995, A mechanism for syn- collisional rock exhumation and associated normal faulting: Results from physical modelling: Earth and Planetary Science Letters, v. 132, no. 1-4, p. 225-232, doi:10.1016/0012-821X(95)00042-B.

Chew, D.M., and Donelick, R.A., 2012, Combined apatite fission track and U–Pb dating by LA-ICPMS and its application in apatite provenance analysis, in Sylvester, P., ed., Quantitative Mineralogy and Microanalysis of Sediments and Sedimentary Rocks: Mineralogical Association of Canada Short Course Series 42, p. 219–247.

Chew, D.M., Donelick, R.A., Donelick, M.B., Kamber, B.S., and Stock, M., 2014a, Apatite chlorine concentration measurements by LA-ICP-MS: Geostandards and Geoanalytical Research, v. 38, no. 1, p. 23–35, doi:10.1111/j.1751- 908X.2013.00246.x.

Chew, D.M., Petrus, J.A., and Kamber, B.S., 2014b, U–Pb LA-ICPMS dating using accessory mineral standards with variable common Pb: Chemical Geology, v. 363, p. 185–199.

Clift, P.D., and Blusztajn, J., 2005, Reorganization of the western Himalayan river system after five million years ago: Nature, v. 438, p. 1001-1003, doi:10.1038/nature04379.

Coleman, R.G., Fleck, R.J., Hedge, C.E., and Ghent, E.D., 1977, The volcanic rocks of southwest Saudi Arabia and the opening of the Red Sea: Saudi Arabian Directorate General of Mineral Resources Bulletin, v. 22, p. D1-D30.

Coleman, R.G., Gregory, R.T., and Brown, G.F., 1983, Cenozoic volcanic rocks of Saudi Arabia. U.S. Geological Survey Open-File Report 83-788, 82 p.

Courtillot, V., Jaupart, C., Manighetti, I., Tapponnier, P., and Besse, J., 1999, On causal links between flood basalts and continental breakup: Earth and Planetary Science Letters, v. 166, no. 3-4, p. 177–195, doi:10.1016/S0012-821X(98)00282-9. d’Almeida, G.A.F., 2010, Structural Evolution History of the Red Sea Rift: Geotectonics, v. 44, no. 3, p. 271-282.

Davison, I., Al-Kadasi, M., Al-Khirbash, S., Al-Subbary, A.K., Baker, J., Blakey, S., Bosence, D., Dart, C., Heaton, R., McClay, K., Menzies, M., Nichols, G., Owen, L., and Yelland, A., 1994, Geological evolution of the southeastern Red Sea Rift margin, Republic of Yemen: Geological Society of America Bulletin, v. 106, p. 1474–1493. 78

Turab, S.A., 2016

Ding, L., Qasim, M., Jadoon, I.A.K., Khan, M.A., Xu, Q., Cai, F., Wang, H., Baral, U., and Yue, Y., 2016, The India–Asia collision in north Pakistan: Insight from the U–Pb detrital zircon provenance of Cenozoic foreland basin: Earth and Planetary Science Letters, v. 455, p. 49-61.

DiPietro, J.A., and Pogue, K.R., 2004, Tectonostratigraphic subdivisions of the Himalaya: A view from the west: Tectonics, v. 23, TC5001, 20 p., doi:10.1029/2003TC001554.

Dobson, K.J., Stuart, F.M., Dempster, T.J., and EIMF, 2008, U and Th zonation in Fish Canyon Tuff zircons: Implications for a zircon (U–Th)/He standard: Geochimica et Cosmochimica Acta, v. 72, no. 19, p. 4745-4755, doi:10.1016/j.gca.2008.07.015.

Donelick, R.A., O’Sullivan, P.B., and Ketcham, R.A., 2005, Apatite fission-track analysis, in Reiners, P.W., and Ehlers, T.E., eds., Low-temperature Thermochronology: Techniques, Interpretations, and Applications: Reviews in Mineralogy and Geochemistry, v. 58, p. 49–94.

England, P., 1981, Metamorphic pressure estimates and sediment volumes for the Alpine orogeny: an independent control on geobarometers?: Earth and Planetary Science Letters, v. 56, p. 387-397, doi:10.1016/0012-821X(81)90142-4.

Farley, K.A., 2002, (U-Th)/He dating: Techniques, calibrations, and applications, in Porcelli, D.P., Ballentine, C.J., and Wieler, R., eds., Noble gases in geochemistry and cosmochemistry: Reviews in Mineralogy and Geochemistry, v. 47, p. 819–844, doi: 10.2138/rmg.2002.47.18.

Farley, K.A., Wolf, R.A., and Silver, L.T., 1996, The effects of long alpha-stopping distances on (U-Th)/He ages: Geochimica et Cosmochimica Acta, v. 60, no. 21, p. 4223-4229, doi:10.1016/S0016-7037(96)00193-7.

Flowers, R.M., Ketcham, R.A., Shuster, D.L., and Farley, K.A., 2009, Apatite (U–Th)/He thermochronometry using a radiation damage accumulation and annealing model: Geochimica et Cosmochimica Acta, v. 73, p. 2347–2365.

Foeken, J.P.T., Stuart, F.M., Dobson, K.J., Persano, C., and Vilbert, D., 2006, A diode laser system for heating minerals for (U–Th)/He chronometry: Geochemistry, Geophysics, Geosystems (G3), v. 7, no. 4, Q04015, 9 p., doi:10.1029/2005GC001190.

79

Turab, S.A., 2016

Fontan, D., Schouppe, M., Hunziker, C.J., Martinotti, G., and Verkaeren, J., 2000, Metamorphic evolution, 40Ar-39Ar chronology and tectonic model for the Neelum valley, Azad Kashmir, NE Pakistan, in Khan, M.A, Treloar, P.J., Searle, M.P., and Jan, M.Q., eds., Tectonics of the Nanga Parbat Syntaxis and the Western Himalaya: Geological Society, London, Special Publication, v. 170, p. 431-453, doi:10.1144/GSL.SP.2000.170.01.23.

Galbraith, R.F., 1981. On statistical models for fission track counts. Mathematical Geology, v. 13, no. 6, p. 471–478.

Gallagher, K., 2012, Transdimensional inverse thermal history modeling for quantitative thermochronology: Journal of Geophysical Research, v. 117, B02408, 16 p., doi:10.1029/2011JB008825.

Gallagher, K., Brown, R., and Johnson, C., 1998, Fission track analysis and its applications to geological problems: Annual Review of Earth and Planetary Sciences, v. 26, p. 519– 572, doi:10.1146/annurev.earth.26.1.519.

Garcia-Castellanos, D., and Jiménez-Munt, I., 2015, Topographic Evolution and Climate Aridification during Continental Collision: Insights from Computer Simulations: PLoS ONE, v. 10, no. 8, e0132252, 32 p., doi:10.1371/journal.pone.0132252.

Garfunkel, Z., 1988, Relation between continental rifting and uplifting: evidence from the Suez rift and northern Red Sea: Tectonophysics, v. 150, no. 1-2, p. 33–49, doi:10.1016/0040-1951(88)90294-6.

Gettings, M.E., Blank, H.R., Jr., Mooney, W.D., and Healey, J.H., 1986, Crustal Structure of Southwestern Saudi Arabia: Journal of Geophysical Research: Solid Earth, v. 91, no. B6, p. 6491–6512, doi:10.1029/JB091iB06p06491.

Ghebreab, W., Carter, A., Hurford, A.J., and Talbot, C.J., 2002, Constraints for timing of extensional tectonics in the western margin of the Red Sea in Eritrea: Earth and Planetary Science Letters, v. 200, p. 107–119.

Gilchrist, A.R., Kooi, H., and Beaumont, C., 1994, Post-Gondwana geomorphic evolution of southwestern Africa: Implications for the controls on landscape development from observations and numerical experiments: Journal of Geophysical Research: Solid Earth, v. 99, no. B6, p. 12211–12228, doi:10.1029/94JB00046.

80

Turab, S.A., 2016

Gleadow, A.J.W., Duddy, I.R., and Lovering, J.F., 1983, Fission track analysis: a new tool for the evaluation of thermal histories and hydrocarbon potential: Australian Petroleum Exploration Association Journal, v. 23, no. 1, p. 93-102.

Greco, A., and Spencer, D.A., 1993, A section through the Indian Plate, Kaghan Valley, NW Himalaya, Pakistan, in Treloar, P.J., and Searle, M.P., eds., Himalayan Tectonics: Geological Society, London, Special Publications, v. 74, p. 221-236, doi:10.1144/GSL.SP.1993.074.01.16.

Guenthner, W.R., Reiners, P.W., Ketcham, R.A., Nasdala, L., and Giester, G., 2013, Helium diffusion in natural zircon: Radiation damage, anisotropy, and the interpretation of zircon (U-Th)/He thermochronology: American Journal of Science, v. 313, p. 145- 198, doi:10.2475/03.2013.01.

Guillot, S., Mahéo, G., de Sigoyer, J., Hattori, K.H., and Pêcher, A., 2008, Tethyan and Indian subduction viewed from the Himalayan high- to ultrahigh-pressure metamorphic rocks: Tectonophysics, v. 451, no. 1-4, p. 225-241, doi:10.1016/j.tecto.2007.11.059.

Hergarten, S., Robl, J., and Stüwe, K., 2014, Extracting topographic swath profiles across curved geomorphic features: Earth Surface Dynamics, v. 2, p. 97-104, doi:10.5194/esurf-2-97-2014.

Hill, R.I., 1991, Starting plumes and continental break-up: Earth and Planetary Science Letters, v. 104, no. 2-4, p. 398–416, doi:10.1016/0012-821X(91)90218-7.

Hodges, K.V., Parrish, R.R., Housh, T.B., Lux, D.R., Burchfield, B.C., Royden, L.H., and Chen, Z., 1992, Simultaneous Miocene extension and shortening in the Himalayan Orogen: Science, v. 258, p. 1466–1470.

Hodges, K.V., Wobus, C., Ruhl, K., Schildgen, T., and Whipple, K., 2004, Quaternary deformation, river steepening, and heavy precipitation at the front of the Higher Himalayan ranges: Earth and Planetary Science Letters, v. 220, p. 379–389, doi:10.1016/S0012-821X(04)00063-9.

Issawi, B., Francis, M., El Hinnawy, M., Mehanna, A., and El Deftar, T., 1971, Geology of Safaga-Quseir coastal plain and of Mohammed Rabah area: Annals of the Geological Survey of Egypt, v. 1, p. 1-20.

81

Turab, S.A., 2016

Kazmi, A.H., and Jan, M.Q., 1997, Geology and Tectonics of Pakistan: Graphic Publishers, Karachi, 554 p.

Ketcham, R.A., Carter, A., Donelick, R.A., Barbarand, J., and Hurford, A.J., 2007, Improved modeling of fission-track annealing in apatite: American Mineralogist, v. 92, p. 799– 810, doi:10.2138/am.2007.2281.

Ketcham, R.A., Gautheron, C., and Tassan-Got, L., 2011, Accounting for long alpha-particle stopping distances in (U–Th–Sm)/He geochronology: Refinement of the baseline case: Geochimica et Cosmochimica Acta, v. 75, no. 24, p. 7779-7791, doi:10.1016/j.gca.2011.10.011.

Khan, M.A., Bendick, R., Bhat, M.I., Bilham, R., Kakar, D.M., Khan, S.F., Lodi, S.H., Qazi, M.S., Singh, B., Szeliga, W., and Wahab, A., 2008, Preliminary geodetic constraints on plate boundary deformation on the western edge of the Indian plate from TriGGnet (Tri-University GPS Geodesy Network): Journal of Himalayan Earth Sciences, v. 41, p. 71-87.

Kirstein, L.A., Sinclair, H., Stuart, F.M., and Dobson, K., 2006, Rapid early Miocene exhumation of the Ladakh batholith, western Himalaya: Geology, v. 34, no. 12, p. 1049–1052, doi:10.1130/G22857A.1.

Kohn, B.P., and Eyal, M., 1981, History of uplift of the crystalline basement of Sinai and its relation to opening of the Red Sea as revealed by fission track dating of apatites: Earth and Planetary Science Letters, v. 52, no. 1, p. 129-141, doi:10.1016/0012- 821X(81)90215-6.

Kooi, H., and Beaumont, C., 1994, Escarpment evolution on high-elevation rifted margins: insights derived from a surface processes model that combines diffusion, advection, and reaction: Journal of Geophysical Research: Solid Earth, v. 99, no. B6, p. 12191– 12209, doi:10.1029/94JB00047.

Koons, P.O., Zeitler, P.K., and Hallet, B., 2013, Tectonic aneurysms and mountain building, in Shroder, J. (Editor in Chief), and Owen, L.A. (Ed.), Treatise on Geomorphology: Academic Press, San Diego, CA, v. 5, Tectonic Geomorphology, p. 318–349.

82

Turab, S.A., 2016

Laslett, G.M., Green, P.F., Duddy, I.R., and Gleadow, A.J.W., 1987, Thermal annealing of fission tracks in apatite 2. A quantitative analysis: Chemical Geology (Isotope Geoscience Section), v. 65, p. 1-13.

Le Fort, P., Debon, F., and Sonet, J., 1980, The “Lesser Himalayan” cordierite granite belt. typology and age of the pluton of Mansehra (Pakistan), in Proceedings of the International Committee of Geodynamics, Group 6, Peshawar: Geological Bulletin University of Peshawar (Special Issue), v. 13, p. 51-61.

Lister, G.S., and Davis, G.A., 1989, The origin of metamorphic core complexes and detachment faults formed during Tertiary continental extension in the northern Colorado River region, U.S.A.: Journal of Structural Geology, v. 11, no. 1-2, p. 65- 94, doi:10.1016/0191-8141(89)90036-9.

Madden, G.T., Schmidt, D.L., and Whitmore, F.C., Jr., 1983, Mastritherium (Artiodactyla, Anthracotheriidae) from Wadi Sabya, southwestern Saudi Arabia: An earliest Miocene age for continental rift-valley volcanic deposits of the Red Sea margin: U.S. Geological Survey Open-File Report 83-488, 24 p.

Makris, J., Tsironidis, J., and Richter, H., 1991a, Heatflow density distribution in the Red Sea: Tectonophysics, v. 198, no. 2-4, p. 383-393, doi:10.1016/0040-1951(91)90162-L.

Makris, J., Henke, C.H., Egloff, F., and Akamaluk, T., 1991b, The gravity field of the Red Sea and East Africa: Tectonophysics, v. 198, no. 2-4, p. 369-381, doi:10.1016/0040- 1951(91)90161-K.

Malik, R.H., Schouppe, M., Fontan, D., Verkaeren, J., Martinotti, G., Ahmed, Kh.S., and Qureshi, S., 1996, Geology of the Neelum valley, district Muzaffarabad, Azad Kashmir, Pakistan: Geological Bulletin University of Peshawar, v. 29, p. 91-111.

Matmon, A., Bierman, P., and Enzel, Y., 2002, Pattern and tempo of great escarpment erosion: Geology, v. 30, no. 12, p. 1135-1138, doi:10.1130/0091- 7613(2002)030<1135:PATOGE>2.0.CO;2.

Meesters, A.G.C.A., and Dunai, T.J., 2005, A non-iterative solution of the (U-Th)/He age equation: Geochemistry, Geophysics, Geosystems (G3), v. 6, no. 4, Q04002, 3 p., doi:10.1029/2004GC000834.

83

Turab, S.A., 2016

Meigs, A.J., Burbank, D.W., and Beck, R.A., 1995, Middle-late Miocene (>10 Ma) formation of the Main Boundary thrust in the western Himalaya: Geology, v. 23, p. 423-426, doi:10.1130/0091-7613(1995)023<0423:MLMMFO>2.3.CO;2.

Menzies, M., Gallagher, K., Yelland, A., and Hurford, A.J., 1997, Volcanic and non-volcanic rifted margins of the Red Sea and Gulf of Aden: Crustal cooling and margin evolution in Yemen: Geochimica et Cosmochimica Acta, v. 61, no. 12, p. 2511– 2527, doi:10.1016/S0016-7037(97)00108-7.

Molnar, P., and England, P., 1990, Late Cenozoic uplift of mountain ranges and global climate change: chicken or egg?: Nature, v. 346, p. 29-34.

Mooney, W.D., Gettings, M.E., Blank, H.R., and Healey, J.H., 1985, Saudi Arabian seismic- refraction profile: a traveltime interpretation of crustal and upper mantle structure: Tectonophysics, v. 111, no. 3-4, p. 173–246.

Mouthereau, F., Lacombe, O., and Verges, J., 2012, Building the Zagros collisional orogen: Timing, strain distribution and the dynamics of Arabia/Eurasia plate convergence: Tectonophysics, v. 532-535, p. 27-60, doi:10.1016/j.tecto.2012.01.022.

Omar, G.I., and Steckler, M.S., 1995, Fission track evidence on the initial rifting of the Red Sea: Two pulses, no propagation: Science, v. 270, no. 5240, p. 1341–1344, doi:10.1126/science.270.5240.1341.

Omar, G.I., Kohn, B.P., Lutz, T.M., and Faul, H., 1987, The cooling history of Silurian to Cretaceous alkaline ring complexes, south Eastern Desert, Egypt, as revealed by fission-track analysis: Earth and Planetary Science Letters, v. 83, no. 1-4, p. 94-108, doi:10.1016/0012-821X(87)90054-9.

Omar, G.I., Steckler, M.S., Buck, W.R., and Kohn, B.P., 1989, Fission-track analysis of basement apatites at the western margin of the Gulf of Suez rift, Egypt: evidence for synchroneity of uplift and subsidence: Earth and Planetary Science Letters, v. 94, no. 3-4, p. 316–328, doi:10.1016/0012-821X(89)90149-0.

Pallister, J.S., 1987, Magmatic history of Red Sea rifting: Perspective from the central Saudi Arabian coastal plain: Geological Society of America Bulletin, v. 98, no. 4, p. 400- 417, doi:10.1130/0016-7606(1987)98<400:MHORSR>2.0.CO;2.

84

Turab, S.A., 2016

Patriat, P., and Achache, J., 1984, India-Eurasia collision chronology has implications for crustal shortening and driving mechanism of plates: Nature, v. 311, p. 615-621.

Pêcher, A., Giuliani, G., Garnier, V., Maluski, H., Kausar, A.B., Malik, R.H., and Muntaz, H.R., 2002, Geology, geochemistry and Ar–Ar geochronology of the Nangimali ruby deposit, Nanga Parbat Himalaya (Azad Kashmir, Pakistan): Journal of Asian Earth Sciences, v. 21, p. 265-282.

Pêcher, A., Seeber, L., Guillot, S., Jouanne, F., Kausar, A., Latif, M., Majid, A., Maheo, G., Mugnier, J.L., Rolland, Y., van der Beek, P., and Van Melle, J., 2008, Stress field evolution in the northwest Himalayan syntaxis, northern Pakistan: Tectonics, v. 27, TC6005, 26 p, doi:10.1029/2007TC002252.

Persano, C., Stuart, F.M., Bishop, P., and Barfod, D.N., 2002, Apatite (U–Th)/He age constraints on the development of the Great Escarpment on the southeastern Australian passive margin: Earth and Planetary Science Letters, v. 200, no. 1–2, p. 79-90, doi:10.1016/S0012-821X(02)00614-3.

Reiners, P.W., 2005, Zircon (U-Th)/He Thermochronometry, in Reiners, P.W., and Ehlers, T.E., eds., Low-temperature Thermochronology: Techniques, Interpretations, and Applications: Reviews in Mineralogy and Geochemistry, v. 58, p. 151-179.

Ring, U., Brandon, M.T., Willett, S.D., and Lister, G.S., 1999, Exhumation processes, in Ring, U., Brandon, M.T., Willett, S.D., Lister, G.S., eds., Exhumation processes: normal faulting, ductile flow and erosion: Geological Society, London, Special Publications, v. 154, p. 1-27.

Rolland, Y., 2002, From intra-oceanic convergence to post-collisional evolution: example of the India-Asia convergence in NW Himalaya, from Cretaceous to present, in Rosenbaum, G., and Lister, G.S., eds., Reconstruction of the evolution of the Alpine- Himalayan Orogen: Journal of the Virtual Explorer, v. 8, p. 193-216, doi:10.3809/jvirtex.2002.00052.

Sahagian, D., 1988, EPeirogenic motions of Africa as inferred from Cretaceous shoreline deposits: Tectonics, v. 7, no. 1, p. 125-138, doi:10.1029/TC007i001p00125.

Sayab, M., and Khan, M.A., 2010, Temporal evolution of surface rupture deduced from coseismic multi-mode secondary fractures: insights from the October 8, 2005 (Mw

85

Turab, S.A., 2016

7.6) Kashmir earthquake, NW Himalaya: Tectonophysics, v. 493, p. 58-73, doi:10.1016/j.tecto.2010.07.001.

Schmidt, D.L., Hadley, D.G., and Brown, G.F., 1983, Middle Tertiary continental rift and evolution of the Red Sea in southwestern Saudi Arabia: U.S. Geological Survey Open-File Report 83-641, 56 p.

Schneider, D.A., Zeitler, P.K., Kidd, W.S.F., and Edwards, M.A., 2001, Geochronologic constraints on the tectonic evolution and exhumation of Nanga Parbat, western Himalaya syntaxis, revisited: The Journal of Geology, v. 109, p. 563–583.

Searle, M.P., Law, R.D., Godin, L., Larson, K.P., Streule, M.J., Cottle, J.M., and Jessup, M.J., 2008, Defining the Himalayan Main Central Thrust in Nepal: Journal of the Geological Society, London, v. 165, p. 523–534, doi:10.1144/0016-76492007-081.

Seeber, L., and Pêcher, A., 1998, Strain partitioning along the Himalayan arc and the Nanga Parbat antiform: Geology, v. 26, p. 791-794, doi:10.1130/0091- 7613(1998)026<0791:SPATHA>2.3.CO;2.

Sengör, A.M.C., and Burke, K., 1978, Relative timing of rifting and volcanism on Earth and its tectonic implications: Geophysical Research Letters, v. 5, no. 6, p. 419-421, doi:10.1029/GL005i006p00419.

Shah, S.M.I., 2009, Stratigraphy of Pakistan: Geological Survey of Pakistan Memoirs, v. 22, 381 p.

Shuster, D.L., Flowers, R.M., and Farley, K.A., 2006, The influence of natural radiation damage on helium diffusion kinetics in apatite: Earth and Planetary Science Letters, v. 249, no. 3-4, p. 148–161, doi:10.1016/j.epsl.2006.07.028.

Stacey, J.S., and Kramers, J.D., 1975, Approximation of terrestrial lead isotope evolution by a two-stage model: Earth and Planetary Science Letters, v. 26, p. 207-221.

Steckler, M.S., and Omar, G.I., 1994, Controls on erosional retreat of the uplifted rift flanks at the Gulf of Suez and northern Red Sea: Journal of Geophysical Research: Solid Earth, v. 99, no. B6, p. 12159–12173, doi:10.1029/94JB00278.

86

Turab, S.A., 2016

Stern, R.J., and Johnson, P., 2010, Continental lithosphere of the Arabian Plate: A geologic, petrologic, and geophysical synthesis: Earth-Science Reviews, v. 101, no. 1-2, p. 29– 67, doi:10.1016/j.earscirev.2010.01.002.

Stüwe, K., 1991, Flexural constraints on the denudation of asymmetric mountain belts: Journal of Geophysical Research: Solid Earth, v. 96, no. B6, p. 10401-10408, doi:10.1029/91JB00679.

Summerfield, M.A., 1989, Tectonic geomorphology: convergent plate boundaries, passive continental margins and supercontinent cycles: Progress in Physical Geography, v. 13, no. 3, p. 431-441, doi:10.1177/030913338901300306.

Summerfield, M.A., ed., 2000, Geomorphology and Global Tectonics: Wiley, Chichester.

Treloar, P.J., Coward, M.P., Chambers, A.F., Izatt, C.N., and Jackson, K.C., 1992, Thrust geometries, interferences and rotations in the Northwest Himalaya, in McClay, K.R., ed., Thrust Tectonics: Chapman and Hall, London, UK, p. 325–342.

Treloar, P.J., Potts, G.J., Wheeler, J., and Rex, D.C., 1991, Structural evolution and asymmetric uplift of the Nanga Parbat syntaxis, Pakistan Himalaya: Geologische Rundschau, v. 80, no. 2, p. 411–428.

Treloar, P.J., Rex, D.C., Guise, P.G., Coward, M.P., Searle, M.P., Windley, B.F., Petterson, M.G., Jan, M.Q., and Luff, I.W., 1989, K-Ar and Ar-Ar geochronology of the Himalayan collision in NW Pakistan: constraints on the timing of suturing, deformation, metamorphism and uplift: Tectonics, v. 8, no. 4, p. 881-909.

Treloar, P.J., Rex, D.C., Guise, P.G., Wheeler, J., Hurford, A.J., and Carter, A., 2000, Geochronological constraints on the evolution of the Nanga Parbat syntaxis, Pakistan Himalaya, in Khan, M.A., Treloar, P.J., Searle, M.P., and Jan, M.Q., eds., Tectonics of the Nanga Parbat Syntaxis and the Western Himalaya: Geological Society, London, Special Publication, v. 170, p. 137-162, doi:10.1144/GSL.SP.2000.170.01.08.

Turab, S.A., 2013, Structural geology of the surroundings of muzaffarabad with emphasis on neotectonics: Masters of Philosophy dissertation, National Centre of Excellence in Geology University of Peshawar, 103 p.

87

Turab, S.A., 2016

Van der Beek, P., Van Melle, J., Guillot, S., Pêcher, A., Reiners, P.W., Nicolescu, S., and Latif, M., 2009, Eocene Tibetan plateau remnants preserved in the northwest Himalaya: Nature Geoscience, v. 2, p. 364-368, doi:10.1038/NGEO503.

Vannay, J-C., Grasemann, B., Rahn, M., Frank, W., Carter, A., Baudraz, V., and Cosca, M., 2004, Miocene to Holocene exhumation of metamorphic crustal wedges in the NW Himalaya: Evidence for tectonic extrusion coupled to fluvial erosion: Tectonics, v. 23, TC1014, 24 p., doi:10.1029/2002TC001429.

Voggenreiter, W., and Hötzl, H., 1989, Kinematic evolution of the southwestern Arabian continental margin: implications for the origin of the Red Sea: Journal of African Earth Sciences, v. 8, no. 2–4, p. 541-564, doi:10.1016/S0899-5362(89)80044-2.

Whipp, D.M., Jr., Beaumont, C., and Braun, J., 2014, Feeding the “aneurysm”: Orogen- parallel mass transport into Nanga Parbat and the western Himalayan syntaxis: Journal of Geophysical Research: Solid Earth, v. 119, p. 5077–5096, doi:10.1002/2013JB010929.

Whipple, K., Wobus, C., Crosby, B., Kirby, E., and Sheehan, D., 2007, New Tools for Quantitative Geomorphology: Extraction and Interpretation of Stream Profiles from Digital Topographic Data: Geological Society of America Annual Meeting, Boulder, CO., 26 p., (http://www.geomorphtools.org/Tools/StPro/StPro.htm) (accessed October 2016).

Whittaker, A.C., 2012, How do landscapes record tectonics and climate?: Lithosphere, v. 4, no. 2, p. 160-164, doi: 10.1130/RF.L003.1.

Wilke, F.D.H., Sobel, E.R., O’Brien, P.J., and Stockli, D.F., 2012, Apatite fission track and (U–Th)/He ages from the Higher Himalayan Crystallines, Kaghan Valley, Pakistan: Implications for an Eocene Plateau and Oligocene to Pliocene exhumation: Journal of Asian Earth Sciences, v. 59, p. 14-23.

Wobus, C., Whipple, K.X., Kirby, E., Snyder, N., Johnson, J., Spyropolou, K., Crosby, B., and Sheehan, D., 2006, Tectonics from topography: Procedures, promise, and pitfalls, in Willet, S.D., Hovius, N., Brandon, M.T., and Fisher, D.M., eds., Tectonics, Climate, and Landscape Evolution: Geological Society of America, Special Paper 398, Penrose Conference Series, p. 55-74, doi:10.1130/2006.2398(04).

88

Turab, S.A., 2016

Zeitler, P.K., 1985, Cooling history of the NW Himalaya, Pakistan: Tectonics, v. 4, no. 1, p. 127-151.

Zeitler, P.K., Johnson, N.M., Naeser, C.W., and Tahirkheli, R.A.K., 1982, Fission-tack evidence for Quaternary uplift of the Nanga Parbat region, Pakistan: Nature, v. 298, p. 255-257.

Zeitler, P.K., Koons, P.O., Bishop, M.P., Chamberlain, C.P., Craw, D., Edwards, M.A., Hamidullah, S., Jan, M.Q., Khan, M.A., Khattak, M.U.K., Kidd, W.S.F., Mackie, R.L., Meltzer, A.S., Park, S.K., Pêcher, A., Poage, M.A., Sarker, G., Schneider, D.A., Seeber, L., and Shroder, J.F., 2001a, Crustal reworking at Nanga Parbat, Pakistan: Metamorphic consequences of thermal-mechanical coupling facilitated by erosion: Tectonics, v. 20, no. 5, p. 712-728.

Zeitler, P.K., Meltzer, A.S., Koons, P.O., Craw, D., Hallet, B., Chamberlain, C.P., Kidd, W.S.F., Park, S.K., Seeber, L., Bishop, M., and Shroder, J., 2001b, Erosion, Himalayan geodynamics, and the geomorphology of metamorphism: GSA Today, v. 11, no. 1, p. 4-9.

89