Glacial Geology of Bayan Har Shan, Northeastern Tibetan Plateau

Glacial geology of Bayan Har Shan, northeastern Tibetan Plateau

Jakob Heyman

Abstract

The paleoglaciology of the Tibetan Plateau is still largely unexplored, despite its importance for regional and global climate reconstructions. In this thesis a comprehensive glacial geological record is presented from an extensive part of the northeastern Tibetan Plateau centred on the Bayan Har Shan. Glacial reconstructions for this region range from restricted mountain glaciers through the intermediate-size regional-scale Huang He ice sheet to a plateau-scale Tibetan ice sheet. To provide a robust basis for glacial reconstructions, this thesis provides conclusions based on two principle methods, remote sensing and field studies. The remote sensing of a 90 m resolution digital elevation model and 15- and 30 m resolution satellite imagery renders a detailed data set with complete spatial coverage of large- and medium-scale glacial landforms, and large-scale plateau geomorphology. Observations from fieldwork campaigns add detailed point information for the distribution of glacial deposits. Geomorphological glacial traces such as glacial valleys, glacial lineations, marginal moraines, meltwater channels, and hummocky terrain occur frequently in elevated mountain areas, indicating former alpine-style glaciations. Glacial deposits in the form of till, erratic boulders, and glaciofluvial sediments are common in areas with mapped glacial landforms, but also beyond, in areas lacking large-scale glacial landforms. For extensive plateau areas in-between formerly glaciated mountain blocks, there is a striking absence of glacial landforms and sediments, indicating that these areas, perhaps, never were ice covered. Interestingly, glacial deposits occur further away from the mountain blocks than the large- and medium-scale glacial landforms, indicating insignificant erosion beneath the maximum ice covers close to their margins.

The large-scale geomorphology of the northeastern Tibetan Plateau is characterised by a low-relief plateau surface with glacial valleys in elevated mountain blocks and marginal steep V-shaped valleys. This geographical distribution indicates a dominance of glacial erosion in the elevated mountain areas and a dominance of fluvial erosion along the steep plateau margins, dissecting a relict plateau surface. The outline of the relict plateau surface mimics the proposed outline of the Huang He ice sheet, suggesting that the inferred ice sheet may represent a misinterpreted relict surface with scattered glacial traces.

In conclusion, the glacial geology examined in the Bayan Har Shan region is consistent with paleo- glaciers of varying extent restricted to elevated mountain areas. Even though extensive icefields/ice caps were centred on discrete mountain areas, there is no indication that these ice masses merged but rather that they were separated from each other by unglaciated plateau areas. The presented glacial geological record will be used in further studies towards a robust paleoglaciological reconstruction for the northeastern Tibetan Plateau.

Cover photo: Erratic granite boulder north of central Bayan Har Shan with a U-shaped cross-section of the parent glacial valley in the background.

This thesis consists of a summary, three papers and a map:

Paper I, including a map on attached CD (P I)

Heyman J, Hättestrand C, Stroeven AP. in press: Glacial geomorphology of the Bayan Har sector of the NE Tibetan plateau. Journal of Maps .

Paper II (P II)

Stroeven AP, Hättestrand C, Heyman J, Harbor J, Li YK, Zhou LP, Caffee MW, Alexanderson H, Kleman J, Ma HZ, Liu GN. in press: Landscape analysis of the Huang He headwaters, NE Tibetan Plateau – patterns of glacial and fluvial erosion. Geomorphology .

Paper III (P III)

Heyman J, Stroeven AP, Alexanderson H, Hättestrand C, Li YK, Harbor J, Caffee MW, Zhou LP, Veres D, Liu F, Machiedo M. manuscript: Field data for the glacial extent in Bayan Har Shan, NE Tibetan Plateau – paleo-glaciers restricted to separate mountain blocks.

For P I, I performed the mapping, building on initial mapping by Clas Hättestrand and Arjen Stroeven, and I wrote most of the text and drew the figures, with significant input from my co- authors. For P II, Clas Hättestrand and Arjen Stroeven conducted the mapping and wrote most of the text. I contributed significantly with data analysis, writing and illustrations. For P III, I supervised and performed most of the data collection, and I was responsible for data analysis, writing and illustrations, with significant input from my co-authors.

2 Introduction areas, as exemplified by the Li et al. (1991) glacial reconstruction for the Tibetan Plateau (cf. Fig. 1). The Tibetan Plateau is a topographic feature of extra- However, because there is an absence of glacial ordinary dimension and, consequently, it has an geological data for extensive parts of the plateau (c.f. P important impact on regional and global climates. It I), presented plateau-wide reconstructions are at best has been proposed, for example, that the uplift of the crude estimates. There is, consequently, a frequently plateau, in response to the Indian-Asian continental broadcasted need for further detailed studies of the collision, caused the Asian monsoon dynamics and a Tibetan glacial geology (e.g. Osmaston 1989; global cooling trend which triggered the onsets of Derbyshire et al. 1991; Rutter 1995; Zheng and Rutter Quaternary climate oscillations (e.g. Molnar and 1998; Benn and Owen 2002; Lehmkuhl and Owen England 1990; Raymo and Ruddiman 1992; An et al. 2005). One of the more intriguing corners of the 2001). However, the glacial history of the Tibetan Tibetan Plateau in this regard is the Bayan Har Shan Plateau, in spite of its proposed significance for climate (Shan means mountain) with the source area of Huang change, is a research field still in its infancy. While He (Yellow River) as it has been credited with variable research on Laurentide and Fennoscandian glaciations paleo-ice extents. This is where the work contained in have at least partly attained a stage of detailed analysis, this thesis has its focus (Fig. 1). research on the Tibetan Plateau glacial history still deals with the basic questions of extent and timing. The upper reaches of Huang He receive their water from the northern slopes of Bayan Har Shan. In this Likely triggered by the suggestion of a plateau-scale area, a regional ice sheet has been proposed to have ice sheet (Kuhle 1985, 2004 and references there-in), existed – the Huang He ice sheet. The Huang He ice the last 20 years have seen a significant increase in sheet, with a peculiar triangular outline and of research on the Tibetan Plateau glacial history. Kuhle Icelandic size, was first described in the (2004 and references there-in), based on inferred palaeoglaciological map by Li et al. (1991) as an glacial geological traces and reconstructed Equilibrium “Estimated extent of paleoglaciation” (Fig. 1). This Line Altitudes (ELA), has suggested that almost the map unit contrasts with the apparently somewhat more entire plateau was covered by an ice sheet of certain “Extent of paleoglaciation” which marks all dimensions similar to the Greenland ice sheet during other formerly glaciated areas, including a more the global Last Glacial Maximum (LGM). This idea restricted glacier extent around Bayan Har Shan that has been subjected to intense disapproval (e.g. falls within the Huang He ice sheet area (Fig. 1). Based Derbyshire et al. 1991; Rutter 1995; Zheng and Rutter mainly on field investigations, Zhou and Li (1998) 1998; Owen et al. , 2005; Lehmkuhl and Owen 2005) presented a glacial reconstruction comprising four and a large amount of data strongly indicates that at glacial stages in the Huang He ice sheet area. Two least during the last few hundred thousand years there stages were identified where glaciers were restricted to has been no plateau-scale ice sheet on the Tibetan the highest mountain areas and these were proposed to Plateau. Two circumstances stand out as particularly correspond to the early and late last glaciation. These problematic for a plateau-scale LGM ice sheet concept. stages were proposed to have been predated by a more First, there is a remarkable absence of glacial extensive marine oxygen isotope stage (MIS) 6 glacial indicators (such as glacial lineations, eskers, ribbed stage and a most extensive Huang He ice sheet MIS 12 moraines, marginal moraines, till, and erratic boulders) glacial stage. However, the physical basis for this for extensive areas of the plateau outside mountain reconstruction is meagre, with only scattered areas. Second, terrestrial cosmogenic nuclide exposure observations of glacial landforms and deposits and an dating and optically stimulated luminescence dating absence of chronological control. Alternative techniques, which are effective chronological tools for interpretations for the glacial history of the Bayan Har glacial geologists (e.g. Gosse and Philips 2001; Shan exist (Lehmkuhl et al. 1998; Zheng and Rutter Bierman 2007), have provided LGM ages from 1998), and these favour restricted glaciations based on restricted mountain areas and a large number of ages an absence of unambiguous glacial traces and they on glacial landforms and deposits older than the late question, therefore, the existence of regional-scale and last glaciation (e.g., Owen et al. 2005). To summarize, plateau-scale ice sheets in this area. the concept of a plateau-scale Tibetan ice sheet during the LGM is generally rejected based on available The Bayan Har Shan investigation area can be dating evidence, and the presence of a unified plateau- described as a low-relief plateau at c. 4300 metres scale ice sheet at any other time before the LGM faces above sea level (m a.s.l.) with elevated mountain areas the current uphill-struggle of an absence of landforms generally reaching up to 1000 m higher, and including and deposits in its support. the significantly higher Anyemaqen Shan (6282 m a.s.l.) in its northeastern corner. Tributaries of Chang The paleoglaciology of the Tibetan Plateau is Jiang (Yangtze River), the major river bordering the dominantly thought of as a collection of paleo-glaciers study area towards the southwest, and Huang He in the of varying extent but restricted to elevated mountain east, are cutting deep V-shaped valleys into the plateau

Fig. 1. The Tibetan Plateau with the study area in the northeastern corner and the interpretation of the Quaternary glacial distribution map (Li et al. 1991).

area. Mountain ranges are generally trending along Remote sensing of the entire study area has been fault lines in a NW-SE orientation. performed using two data sets: the Shuttle Radar Topography Mission (SRTM) 90 m resolution Digital Clearly, the paleoglaciology of the northeastern Elevation Model (DEM) (Jarvis et al. 2006), and Tibetan plateau is not satisfactory researched, with Landsat ETM+ 15 m resolution satellite imagery several crucial questions still unresolved. In my PhD (acquired from Global Land Cover Facility, project I am aiming to provide a robust http://www.landcover.org, accessed 17 January 2008). paleoglaciological reconstruction of the Bayan Har For 3D visualization, Google Earth TM software has Shan area based on mapping, field evidence and been frequently employed. The remote sensing numerical dating techniques. Ideally, a methodology, by virtue of its complete coverage over paleoglaciological reconstruction would provide a extensive areas, provides a powerful tool to yield temporally continuous glacial development for the spatially-continuous records of glacial landforms over study area, including details on glacier outline, large areas and based on uniform information. Its topography, thermal regime, and ice dynamics (cf. predominant limitation is the spatial resolution of the Kleman et al. 2002; Napieralski et al., 2007). However, data, where small-scale features, such as small moraine due to the complexity and difficulties of the glacial ridges and roche moutonnées, may fall outside the inversion problem (Kleman et al. 2006), lower limit of recognition. paleoglaciological reconstructions are more typically composed of two main parts; a spatial dimension Three field campaigns were carried out during 2005, giving the glacier outlines, and a temporal dimension 2006 and 2007. We have investigated landforms and giving the chronology, presenting time slice images of sediments with the aim to either associate features with former glacial extents (e.g. Boulton and Clark 1990; the former presence of a glacier or to exclude glacial Kleman et al. 1997; Svendsen et al. 2004; Ehlers and processes as the formative agent. The field campaigns Gibbard 2007). In this classical way, my thesis deals yielded detailed information on glacial geology, with the spatial dimension of the northeastern Tibetan however, somewhat curtailed by its restricted spatial Plateau paleoglaciology. In particular, it presents the coverage (along roads and walking tracks). input data (in P I, P II, and P III) for a robust reconstruction of the former extent of glaciers. Results

Methodology The results of this thesis are graphically presented in Figure 2, displaying mapped glacial geomorphology Two general methods have been used to map the extent (from P I) from remote sensing and the distribution of of former glaciers: remote sensing and field observed glacial deposits (from P III) from fieldwork, investigations. and are summarized below.

Fig. 2. Glacial geology of Bayan Har Shan.

P I: P II: Heyman J, Hättestrand C, Stroeven AP. in press: Stroeven AP, Hättestrand C, Heyman J, Harbor J, Li Glacial geomorphology of the Bayan Har sector of the YK, Zhou LP, Caffee MW, Alexanderson H, Kleman NE Tibetan plateau. Journal of Maps . J, Ma HZ, Liu GN. in press: Landscape analysis of the Huang He headwaters, NE Tibetan Plateau – patterns We here present the distribution of glacial landforms in of glacial and fluvial erosion. Geomorphology . the Bayan Har Shan study area, including glacial valleys and troughs, glacial lineations, marginal In this paper we present an analysis of the large-scale moraines and moraine remnants, meltwater channels, geomorphology of the Bayan Har Shan study area, and and hummocky terrain. However, there is a notable the relative importance of glacial and fluvial erosion in absence of landform assemblages that are typical for shaping the plateau area. Based on remote sensing, a ice sheets elsewhere on Earth such as glacial lineation tripartite classification system was presented, glacial swarms, ribbed moraines, and eskers. Mapping landscapes (including U-shaped glacially eroded procedures using remote sensing techniques are valleys), relict upland surfaces, and fluvial landscapes described and also include definitions of the mapped dominated by V-shaped valleys formed by fluvial landforms. The glacial landforms are concentrated in incision of tributaries to the Chang Jiang and Huang and around well-defined elevated mountain areas, He. In elevated mountain areas of the relict upland while there is a remarkable absence of glacial surface, the glacial valleys are wider and deeper than landforms for extensive lower-lying plateau areas. In adjacent fluvial valleys, indicating that, integrated over summary, evidence displayed on the map that is part of time, the glacial system has been more erosive than the this publication offers support for former alpine style fluvial system. Along the plateau margin, however, this glaciation but renders no support to the existence of relationship is reversed with dramatic fluvial larger-scale ice sheet glaciation. rejuvenation having consumed/consuming whatever

5 glacial morphologies existed. The outline of the relict beyond that mapped for glacial geomorphologies, upland surface closely mimics the outline of the indicating an absence of large-scale glacial landforms proposed Huang He ice sheet. This circumstance could in areas of most extensive glaciation. In addition, new indicate one of two options with conflicting glacial data shows that previously proposed glacial deposits implications: (1) The Huang He ice sheet was even located in lower-lying plateau areas were probably more extensive because parts of the glacial record are formed by non-glacial processes. In summary, field being consumed by fluvial incision. (2) The proposed evidence support the presence of extensive ice masses regional ice sheet is based on a mis-interpretation of a in the form of separate ice caps or icefields over relict upland surface, whose outline is determined by elevated mountain areas but it leaves no support to the marginal fluvial incision, with scattered glacial traces. presence of any former ice sheet. In the absence of large-scale glacial landforms for extensive areas within the study area, the preferred interpretation is the latter of these. Glacial extent

P III: A preliminary and tentative reconstruction of the Heyman J, Stroeven AP, Alexanderson H, Hättestrand maximum glacial extent permitted by the mapped C, Li YK, Harbor J, Caffee M, Zhou LP, Veres D, Liu glacial geology (Fig. 2, P I, P II, P III) is presented in F, Machiedo M. Manuscript: Field data for the glacial Figure 3. The reconstruction encompasses all glacial extent in Bayan Har Shan, NE Tibetan Plateau – paleo- traces (Fig. 2) and includes a glacial altitude increase glaciers restricted to separate mountain blocks. towards the interior of the plateau (Li et al. 1991; Lehmkuhl 1998; Benn and Owen 1998; P II). The In this paper we present field evidence for glacial reconstructed maximum glacial extent covers 68.000 extent in the Bayan Har Shan and compare glacial km 2, representing an extensive glaciation of geological point observations to the pattern of approximately 50% of the 136.500 km 2 study area. The glaciation inferred from remote sensing data (P I, P II). dominant ice mass was elongated in a NW-SE For the identification and mapping of glacial deposits orientation along Bayan Har Shan. It was, however, we relied on glacial indices such as the presence of significantly smaller than the proposed Huang He ice erratic boulders, of striated clasts, of diamictic sheet of estimated 95.500 km 2 (Li et al. 1991) or sediments, and of clasts of multiple lithologies. Similar 85.500 km 2 (Zhou and Li 1998). to the distribution of glacial geomorphology, glacial deposits are centred on and around elevated mountain areas. However, glacial deposits occur also in a zone

Fig. 3. A preliminary and tentative reconstruction for the maximum glacial extent based on the glacial geological record (cf. Fig. 2). 6 Conclusions represented an icefield or ice cap-type of glaciation.

Based on satellite data of the northeastern Tibetan Plateau and field observations during three consecutive Outlook years, the following conclusions can be drawn: The glacial geological record presented in this thesis is • Glacial landforms occur frequently in elevated an ample record of former glaciations on the mountain areas on the relict plateau surrounding the northeastern Tibetan Plateau. Still, to use this record Bayan Har Shan, indicating former glaciation for a paleoglaciological reconstruction is not a simple strongly controlled by altitude. Distinct sets of and straight-forward task. The glacial geological record marginal moraines and meltwater channels help varies considerably between different areas, potentially distinguishing between discrete glacial extents caused by varying glacial histories and glacier ranging from cirque glaciers to extensive valley dynamics, but also due to the spatial coverage of our glacier networks. There is an absence of glacial field studies. Generally, there is a great amount of landforms for lower-lying intervening plateau areas subjectivity when transforming a glacial geological and ice sheet landform assemblages characteristic record into discrete glacial extents (e.g. Li et al. 1991; for northern Hemisphere mid-altitude ice sheets are Zhou and Li 1998; Fig. 3), leaving the field open for lacking altogether. Thus, glacial geomorphology speculative reconstructions of poor foundation. To get indicates only the former presence of alpine style around this problem, I intend to perform the glaciations. transformation of the glacial geology into glacial extents using a clear methodology based on ELA • Glacial deposits occur frequently in elevated reconstructions. A four point procedure for this mountain areas, similar to the glacial landforms. In transformation is outlined below: addition, glacial deposits occur at lower elevations some distance beyond the mapped glacial 1. Definition of the ELA pattern of the study area. geomorphology. This indicates that icefields which extended beyond the area of visible glacial 2. ELA reconstructions for a part of the study area geomorphology were apparently incapable of with well-defined ice marginal positions. significant glacial erosion, possibly because these where short-lived events. In the lowest-lying plateau 3. Reconstruction of ice marginal positions for the areas, there is a striking absence of glacial deposits. entire study area employing inversion of the ELA The observed distribution of glacial deposits reconstruction (point 2) in combination with the indicates that glaciations were controlled by altitude, ELA pattern (point 1). that the maximum glacial extent exceeds that inferred from glacial geomorphological alone, and 4. Extrapolation of the ice marginal outline between that it offers no support for the presence of paleo-ice the ice marginal locations given by point 3. sheet configurations. The timing of former glaciations in Bayan Har Shan is • The northeastern Tibetan Plateau is composed of a still to be revealed. To constrain the glacial chronology low-relief relict plateau surface with elevated we are performing cosmogenic isotope exposure dating mountain areas. The relict surface is being as well as optically stimulated luminescence dating of consumed by fluvial incision of the Chang Jiang and glacial deposits. We have collected a large number of Huang He tributary rivers. The size and depth of samples from an extensive area, and we intend to adopt glacial valleys testify of glacial erosion surpassing varying cosmogenic analysis techniques, such as dating fluvial erosion in the elevated mountain areas of the of boulder surfaces, sediment depth profiles and sets of plateau surface when integrated over glacial time surface cobbles. In addition, cosmogenic exposure spans. At the plateau margin this relationship is dating of river sediment samples may provide a reversed, with fluvial erosion eradicating whatever measure of erosion rates which can confirm or disprove glacial landforms may have existed. The outline of the idea of a relict plateau surface being consumed by the relict plateau surface broadly resembles the fluvial incision (P II). outline of proposed Huang He ice sheet, indicating that the triangular regional ice sheet may have come about as a misinterpreted relict plateau surface with Acknowledgements scattered glacial traces. First I want to thank my supervisors, Arjen Stroeven, • The glacial geology of the Bayan Har Shan indicates Clas Hättestrand and Helena Alexanderson, who have that the northeastern Tibetan Plateau guided me in a positive, helpful, inspiring and friendly paleoglaciology was restricted to elevated mountain atmosphere. My principle supervisor Arjen has been a areas. In its most expansive outline it probably fantastic support with very valuable comments on 7 various topics, enthusiasm over glacial issues and a courses, conferences and research visits, group dinner-at-University company. Similarly, Clas and meetings and cosmogenic exposure dating (in progress) Helena have been excellent co-supervisors always has been provided by PRIME Lab, Purdue University, ready to help with different issues. Clas was also to a by courtesy of Prof. Marc Caffee, VR/SIDA – Swedish great extent involved in getting me into Research Links grant 348-2004-5684 to Arjen Stroeven paleoglaciological research, as he guided me in such a and the Royal Swedish Academy of Sciences (Margit good way during my undergraduate-degree project on Althins stipendiefond and Hierta-Retzius Swedish marginal moraines. All of my supervisors also stipendiefond), the Swedish Society for Anthropology participated in the Bayan Har fieldwork of 2006, and and Geography (Andréefonden and Vegafonden), Carl have given valuable comments for this summary. I am Mannerfelts fond, A.E.W. Smitts stipendiestiftelse, looking forward to continuing working on this project Ahlmanns fond, Peking University (SU/PKU together with you! Academic Exchange Agreement), C.F. Liljevalch J:ors resestipendium, and Gerard De Geers stiftelse to Jakob This project has involved a large number of helpful Heyman. people who have contributed to this thesis in many different ways. Zhou Liping, at Peking University, has been helping out in a splendid way with the logistics of References my two fieldworks 2006 and 2007. Zhou Liping was also an excellent host for my visit at Peking University An ZS, Kutzbach JE, Prell WL, Porter SC. 2001: after the fieldwork of 2007, providing an open and Evolution of Asian monsoons and phased uplift of the friendly atmosphere and giving me possibilities to Himalayan Tibetan plateau since late Miocene times. conduct research. Liu Gengnian introduced me to the Nature , 411, 62-66. Tibetan Plateau in a very friendly and favourable way, during some memorable days in August 2006 from Benn DI, Owen LA. 1998: The role of the Indian Geermu (Golmud) and across the plateau to Lhasa. Li summer monsoon and the mid-latitude westerlies in Yingkui, Jon Harbor and Marc Caffee have been Himalayan glaciation: review and speculative helping out with several issues, although for sure being discussion. Journal of the Geological Society, London , overwhelmed by other commitments, and they have 155, 353-363. offered very nice company during several meetings and in the field. Jon and Marc have also been extremely Benn DI, Owen LA. 2002: Himalayan glacial helpful and hospitable during my stay at PRIME Lab, sedimentary environments: a framework for Purdue University, for processing TCN samples. Johan reconstructing and dating the former extent of glaciers Kleman has added sharp analysis to the project in an in high mountains. Quaternary International , 97-98, 3- inspiring way, and he has also helped out with practical 25. details. During my two trips to China, two persons have helped out in an extraordinary way, translating, Bierman P. 2007: Cosmogenic glacial dating, 20 years explaining, discussing and helping me with anything I and counting. Geology , 35, 575-576. could ever imagine. Liu Feng during 2007 and Dong Jianyi during 2006 have helped me in a way far Boulton GS, Clark CD. 1990: A highly mobile exceeding what could be expected, and for this I am Laurentide Ice Sheet revealed by satellite images of deeply thankful. In addition to Feng and Jianyi, I have glacial lineations. Nature , 346, 813-817. had very nice company during the fieldworks with Martin Machiedo and Ma Luyi in 2006 and Daniel Derbyshire E, Shi YF, Li JJ, Zheng BX, Li SJ, Wang Veres in 2007, making the sometimes harsh fieldworks JT. 1991: Quaternary glaciation of Tibet: the to very joyful and nice experiences. Göran Alm has geological evidence. Quaternary Science Reviews , 10, been a fantastic resource when it comes to DEM and 485-510. satellite imagery work. Åke Nagrelius at Stockholm University and Li Yun at Peking University have been Ehlers J, Gibbard PL. 2007. The extent and chronology very helpful in connection with my exchange visit to of Cenozoic Global Glaciation. Quaternary Peking University. International , 164-165, 6-20.

I also want to thank fellow PhD students and other staff Gosse JC, Phillips FM. 2001: Terrestrial in situ at the Department of Physical Geography and cosmogenic nuclides: theory and application. Quaternary Geology for creating a very nice Quaternary Science Reviews , 20, 1475-1560. atmosphere to work in. In particular, I want to thank Tim Johnsen for bringing me twice to Jämtland for Jarvis A, Reuter HI, Nelson A, Guevara E. 2006: Hole- fieldwork during the autumn 2005. filled seamless SRTM data V3. International Centre for Tropical Agriculture (CIAT). Available from: Funding for fieldwork, participation in far away PhD http://srtm.csi.cgiar.org [7 January 2008]

8 Kleman J., Hättestrand C., Borgström I., Stroeven A. Osmaston H. 1989: Problems of the Quaternary 1997: Fennoscandian palaeoglaciology reconstructed geomorphology of the Xixabangma region in South using a glacial geological inversion model. Journal of Tibet and Nepal. Zeitschrift für Geomorphologie N.F. Glaciology, 43, 283-299. Supplementband , 76, 147-180.

Kleman J., Hättestrand C., Stroeven AP, Jansson KN, Owen LA, Finkel RC, Barnard PL, Ma HZ, Asahi K, De Angelis H, Borgström I. 2006: Reconstruction of Caffee MW, Derbyshire E. 2005: Climatic and palaeo-ice sheets – inversion of their geomorphological topographic controls on the style and timing of Late record. In Glacier Science and Environmental change , Quaternary glaciation throughout Tibet and the Knight P (ed). Blackwell Publishing, Oxford, 192-198. Himalaya defined by 10 Be cosmogenic radionuclide surface exposure dating. Quaternary Science Reviews , Kuhle M. 1985: Glaciation Research in the Himalayas: 24, 1391-1411. A New Ice Age Theory. Universitas , 27, 281-294. Raymo ME, Ruddiman WF. 1992: Tectonic forcing of Kuhle M. 2004: The high glacial (last ice age and late Cenozoic climate. LGM) ice cover in High and Central Asia. In Nature , 359, 117-122. Quaternary Glaciations - Extent and Chronology, Part

III: South America, Asia, Africa, Australia, Antarctica , Ehlers J, Gibbard PL (eds). Elsevier, Amsterdam, 175- Rutter N. 1995: Problematic ice sheets. Quaternary 199. International , 28, 19-37.

Lehmkuhl F. 1998: Extent and spatial distribution of Svendsen JI, Alexanderson H, Astakhov VI, Demidov Pleistocene glaciations in eastern Tibet. Quaternary I, Dowdeswell JA, Funder S, Gataullin V, Henriksen International , 45/46, 123-134. M, Hjort C, Houmark-Nielsen M, Hubberten HW, Ingólfsson Ó, Jakobsson M, Kjær KH, Larsen E, Lehmkuhl F, Owen LA. 2005: Late Quaternary Lokrantz H, Lunkka JP, Lysa A, Mangerud J, glaciation of Tibet and the bordering mountains: a Matiouchkov A, Murray A, Möller P, Niessen F, review. Boreas , 34, 87-100. Nikolskaya O, Polyak L, Saarnisto M, Siegert C, Siegert MJ, Spielhagen RF, Stein R. 2004: Late Lehmkuhl F, Owen LA, Derbyshire E. 1998: Late Quaternary ice sheet history of northern Eurasia. Quaternary Glacial History of Northeast Tibet. Quaternary Science Reviews , 23, 1229-1271. Quaternary Proceedings , 6, 121-142. Zheng BX. 1989: Controversy Regarding the Existence Li BY, Li JJ, Cui ZJ, Zheng BX, Zhang QS, Wang FB, of a Large Ice Sheet on the Qinghai-Xizang (Tibetan) Zhou SZ, Shi ZH, Jiao KQ, Kang JC. 1991: Plateau during the Quaternary Period. Quaternary Quaternary glacial distribution map of Qinghai-Xizang Research , 32, 121-123. (Tibet) plateau . Science Press, Beijing. (Map scale: 1:3.000.000) Zheng BX, Rutter N. 1998: On the problem of Quaternary glaciations, and the extent and patterns of Molnar P, England P. 1990: Late Cenozoic uplift of Pleistocene ice cover in the Qinghai-Xizang (Tibet) mountain ranges and global climate change: Chicken plateau. Quaternary International , 45/46, 109-122. or egg?. Nature , 346, 29-34. Zhou SZ, Li JJ. 1998: The sequence of Quaternary Napieralski J, Harbor J, Li YK. 2007: Glacial glaciation in the Bayan Har mountains. Quaternary geomorphology and geographic information systems. International , 45/46, 135-142. Earth-Science Reviews , 85, 1-22.