Glacial Hazard Assessment and Risk Management:

Lessons from Tsho Rolpa and New Perspectives

INTRODUCTION

This brief overview has been written to signpost issues relating to glacial hazard assessment and risk management in order to provide a perspective on the science and to some extent the geopolitics involved in such projects. The focus will be the saga of Tsho Rolpa in Rolwaling, Nepal, but I will also include other projects from the Himalayas as necessary.

This article does not purport to be a comprehensive review of glacial hazards – that would take a whole book to achieve. Neither will I necessarily develop arguments here where they have been adequately covered in published literature.

GLACIAL HAZARD TERMONOLOGY

There are various terms that have been used in relation to glacial hazards and these are summarised in Table 1. The term ‘jökulhlaup’ (literally ‘glacier leap’) in glaciological circles is a specific case of a sub-glacial outburst often arising in association with volcanic activity or associated high geothermal heatflux. It relates specifically to the hydrological jacking of overlying ice (Paterson, 1994) as may be caused by increased sub-glacial water pressure from either volcanically-induced rapid melting (as occurs in Iceland, where the term originated) or ice dams. From a physical process perspective, the key process is sub-glacial hydrological jacking of a damming glacier tongue. Consequently, ‘jökulhlaup’ should NOT be used when describing outbursts from glacial lakes such as have occurred in the Himalayas and Peruvian Andes (Reynolds, 1992; Richardson and Reynolds, 2000) as no glaciers are doing any leaping! See also: http://www.swisseduc.ch/glaciers/ glossary/index-en.html.

Table 1: Types of glacier, glacial and related hazards (modified from Reynolds, 1992; Richardson and Reynolds, 2000)

Table 1 describes the various features and their commonly used nomenclature and the time frame associated with each. In this article I will concentrate only on Glacial Lake Outburst Floods (GLOFs) caused from failure of moraine dams. As an aside, I actually dislike this term but it is now widely used in the Himalayan context.

THE TSHO ROLPA HISTORY

Much has been written about Tsho Rolpa (often by people with no knowledge or experience of it) and the recollection of its history is often recorded selectively.

In July 1991 a small glacial lake called Chubung on Ripimo Shar Glacier burst causing damage locally and generating a GLOF downstream that killed one elderly deaf lady in Beding (Reynolds, 1998). The local Sherpas felt that if this small lake could cause this much damage what would happen if Tsho Rolpa were to burst? They wrote a letter to the embassies in Kathmandu pleading for assistance and signed/marked by representatives from each family household. Their plea fell on deaf ears. In 1992 Michel Damen, ITC Enschede, Holland, visited Tsho Rolpa and wrote a brief report on his perception of the situation (Damen, 1992) and in the autumn of that year contacted me and apprised me of the situation as he saw it. In December 2003 I was officially invited by the Water and Energy Commission Secretariat (WECS, His Majesty’s Government of Nepal (HMGN)) in Kathmandu to become involved and help assess the situation, following their own investigations (WECS, 1993; Yamada, 1996). I had previously done a lot of work in the Peruvian Andes on glacial hazard assessment and mitigation with the Peruvian Government (e.g. Reynolds, 1992). It was not until September 1994 that I was able to go physically to Nepal and begin my work at Tsho Rolpa (Reynolds Geo-Sciences Ltd (RGSL), 1994) funded by the Overseas Development Administration, British Government.

The Dutch link through Dr Damen led to involvement with the Netherlands-Nepal Friendship Association that in May 1995 resulted in the first installation of siphons for glacial hazard reduction in the Himalayas (Reynolds, 1999). In November 1995 I worked with WECS to try to develop a national strategy for dealing with glacial hazards that led to WECS developing a proposal for remediating Tsho Rolpa.

In the autumn of 1996 the responsibility of managing glacial hazards was transferred from WECS to the Department of Hydrology and Meteorology (DHM). The WECS GLOF unit was effectively closed down with staff moving elsewhere, including to ICIMOD, and the Japanese workers funded through JICA, who had undertaken several seasons of work at Tsho Rolpa, returning to Japan. This has been the source of the friction between ICIMOD and DHM ever since and was to play a part in the public fracas associated with the 1997 work. More of this later.

To cut a long story short, there was a political hiatus between the Nepali and Dutch governments (WECS and later DHM, Kathmandu, and Ministry of Foreign Affairs, The Hague) in 1995 following the making of a TV film by the Dutch Government apparently without official sanction by the Nepali Government (or payment of the required filming fee). As a consequence, the Nepali side would not speak with the Dutch and vice versa and stalemate existed. However, both sides continued to use me as a go-between, a situation that lasted for 18 months (all pro bono on my part). I felt that if I did not keep the channels of communication open, the situation at Tsho Rolpa could develop to become a potential catastrophe. Geophysical work had been undertaken by OYO Corporation in 1994 (OYO Corporation, 1995) and in 1995 a couple of Dutch students undertook field work in the area (Modder and Van Olden, 1995). In December 2005 HMGN Ministry of Foreign Affairs submitted a proposal to the Royal Netherlands Embassy, New Delhi, requesting assistance to minimise the potential GLOF hazard at Tsho Rolpa. In March 1996 I was able to hold meetings in The Hague with the Dutch Government and submitted a project proposal (RGSL, 1996), which was subsequently approved by HMGN and submitted by them to the Royal Netherlands Government. Following further negotiations, and the publication of Yamada’s report (Yamada, 1996), and further field work (RGSL, 1997) the Dutch Government commissioned a Formulation Mission that eventually occurred in the autumn of 1997 (BPC Hydroconsult, 1997). This was the embryonic start of the ‘Tsho Rolpa GLOF Risk Reduction Project’ that eventually took place in 1999 and 2000. There is also much else that went on behind the scenes geopolitically the telling of which will have to be deferred to another occasion! Of relevance to this e-conference, however, is that much of the original recommendations about the scope of the Tsho Rolpa Project were rejected for what I consider to be political rather than technical or financial reasons. I had tried to introduce the concept of integrated hazard management so that remediation works would produce additional secondary benefits to the local population as part of an integrated rural development plan. The Dutch government sliced whole swathes out of my original suggestions concentrating, rightly or wrongly, solely on the remediation of Tsho Rolpa. One effect is that the local capacity for managing glacial hazards within Nepal is severely limited in terms of not only resources (human and physical) but also in terms of expertise. There are many cases where ‘a little knowledge is a dangerous thing’ and this has exacerbated the problems when dealing with the media in Kathmandu. This continues to be a problem with certain institutions in Kathmandu.

In the summer of 1997 I was dismayed at the panic that ensued concerning the potential failure of the Tsho Rolpa end moraine. Indeed, the situation in May 1997 indicated that failure could be imminent and most probably during the monsoon, when most GLOFs occur. However, it was also stated that it was not clear whether such a failure would occur in the monsoon following or in subsequent years. The science could not give the answer. The popular press took this as meaning that failure would occur in the following few weeks and this prompted the panic. It was not helped by sniping from the sidelines by former WECS staff who were opposed to DHM’s handling of the matter and were holding press conferences that had the effect of undermining DHM’s position. This was ugly and unnecessary, as it was clear to me that the fuel for the media was being added to deliberately by people who had a personal political agenda or who had a grudge to bear and were trying to score points. In actuality, the 1997 monsoon season did turn out to be the most hazardous although, thankfully, no collapse of the moraine dam occurred. Subsequent analysis of the Trakarding Glacier ice front has demonstrated that it goes through a cycle of activity (RGSL, 1999). In some years, the ice cliff is near vertical and produces many ice avalanches and ice calving events that can cause large displacement waves that impact onto the terminal moraine dam. In the monsoon of 1997 two large events occurred, one produced a wave over 1.7 m high and the other produced one the maximum height of which went unrecorded as no one measured the wetted height on the shoreline after the event. Both nearly overtopped the moraine dam but neither had the capacity to cause substantial damage on the distal side of the moraine. At other times in the ice front cycle, the ice cliff becomes much less steep and develops an ice apron in front of it caused by the melting back of the ice at and above the water line. This continues until the ice front becomes near vertical again and the process repeats itself. This has also been observed on other similar glaciers in the Himalayas.

The media frenzy that occurred during the monsoon of 1997 could have been defused had there been an organisation with the authority and scientific gravitas to issue suitable statements to the press. At that time the science behind our understanding of GLOFs and their causes was not well enough developed and the transfer of such knowledge was hampered by the limitations of the project. Even now, despite having a much better understanding of the science of glacial lake behaviour, the media still promotes doom and disaster above scientific reason.

In conclusion, the upshot of the Tsho Rolpa GLOF Risk Reduction Project was the successful installation of an open channel and sluice gates at Tsho Rolpa and the uneventful drawdown of the lake level by 3.5 m by July 2000 (Rana et al., 2000). During the progress of this project substantial amounts of scientific work were undertaken at Tsho Rolpa from 1997 onwards, which will not, for sake of brevity, be chronicled here. DHM continues to monitor the lake and hydro-meteorological data are still being collected. However, there has not been a further glacial hazard assessment of Tsho Rolpa since 2000 and the present status of the terminal moraine dam is unclear. Following detailed geotechnical modelling of the terminal moraine dam a further recommendation was given to DHM suggesting that the lake level should be lowered by a further 11.5 m and preferably by 16.5 m to achieve a maximum reduction in the glacial hazard to achieve internationally recognised Factors of Safety. Recommendations were also made to monitor the situation closely and to undertake periodic reassessments of the glacial hazards to monitor changes in the terminal moraine dam. It is known, for example, that parts of the north-west terminal moraine contain up to 20 m of buried glacier ice that is currently undergoing thermokarstic degeneration. Historically, we know that parts of the moraine can subside very rapidly through ice melt as happened in 1996/1997 when a hole appeared in the moraine to a depth of some 22 m and 55 m across having formed in only a matter of months. An ice cliff was exposed and old glacial structures (crevasses) reactivated leading to very rapid decay. This decreased the effective width of the moraine dam substantially in this area. It is among these reasons that further remediation is considered necessary.

ARUN III

Reference has been made previously to the situation concerning the proposed Arun III Hydro-Electric Power scheme. My involvement was very small but came about after I had inspected aerial photographs taken in 1992 of glacial lakes in the Arun catchment. Staff at WECS revealed this information and asked my opinion. My view at that stage was that if there are glacial lakes in the vicinity of a propose scheme, a proper glacial hazard assessment should be undertaken. I was also able to read the assessment undertaken by the main consultant to the project at that time, who had referred only to maps of the area on which no glacial lakes were indicated. Those maps had been produced in 1985 based on aerial photography from around the mid-1950s prior to the formation of the current glacial lakes. The consultant had concluded that, as there were no glacial lakes present, there was no hazard. I mentioned to the then British Ambassador my concern about the lack of an adequate assessment of the glacial hazards. He had dinner that same night with the Resident Representative of the World Bank in Kathmandu and obviously passed on my concerns. The next I knew was a phone call from the World Bank in Washington DC, USA, inviting me to the meeting of experts in Paris in the spring of 1995. Not being party to the details of the full extent of the Arun III scheme, I had no vested interest in whether the project went ahead or not. My concern was solely the lack of a glacial hazard assessment.

At the meeting I was asked point blank by the Chairman "Considering that this is a $1.2 billion scheme, are glacial hazards enough to kill the project? Yes or no?" My reply was that on the basis of the evidence that I had seen no one could say whether they were or not, as the glacial hazard assessment had not been undertaken adequately. On the basis of this statement, I was asked if I could scope a field investigation to undertake such an assessment. I asked how long I had and was told – "lunch time". With assistance from Drs Jeorg Hanisch and Wolfgang Grabs, who had both worked previously on the GLOF issue in Nepal (Grabs and Hanisch, 1993), we produced a notional scheme costing $328,000. After lunch we were asked for the results of our deliberation and were subsequently granted a sum of $500,000 to be managed by the Federal Institute of Geosciences (BGR, Germany) for whom Dr Hanisch worked. Later that day I was given a message by the head of the German delegation to say that he wanted to see me. I was fearful that I was going to be admonished severely for having so publicly criticised one of Germany’s leading consultancy firms. To the contrary, he congratulated me for having raised the issue so clearly and objectively. However, with two weeks to go before mobilising to the field to undertake the work, the World Bank pulled the project and nothing was done. To this day I know of no professional glacial hazard assessment having been undertaken for this area of Nepal.

This raised the issue of the perception of glacial hazards in the development of hydropower schemes. The perception that glacial hazards exist in a region were seen by some as being sufficient not to venture into a hydropower project. For Nepal, where hydropower development is seen as an important contributor to the economic development of the country, such a negative perspective could be massively devastating economically. As a result of the Paris 1995 meeting the World Bank changed its policy on glacial hazards and since then has insisted that a proper assessment should be undertaken (Reynolds, 1998). This policy then was adopted by UN agencies where they support such development. However, the issue then became – What constitutes a proper glacial hazard assessment? To give an instance, UNOPS in Peru funded a feasibility study for a hydropower scheme in the Peruvian Andes for which a glacial hazard analysis was specified. The problem was that this was all the detail that was given. Potential contractors had to interpret what this meant. In the end the successful bidder had a team that looked at glacial issues but not the actual hazard.

This lack of specification for glacial hazard assessment and the emotive labeling glacial lakes dangerous without reference to any criteria led me to believe that the situation needed to change. This has been enforced by various media misrepresentations, such as the Palcacocha Fiasco in Peru (see below).

THE MEDIA

While the media has enormous power to do good it can also do great harm. Misrepresentation in the media as far as glacial hazards are concerned has had detrimental effects and although it has raised awareness, it has also fuelled the impending doom scenario without justification. Like Jack Ives, I too have been grossly misrepresented in the media, such as by New Scientist. While I firmly believe that glacial hazards are an important issue and one that will undoubtedly increase with continuing climate change, I do not believe that millions of people are at risk or that billions of dollars damage could be caused, despite being (mis-)quoted as saying that I do. Risk management requires the assessment of both hazard and vulnerability. In the Himalayas both are increasing – the hazard, by virtue of growth in the volume of water being stored in glacial lakes and by the possible increase in the likelihood of more GLOFs occurring with greater frequency; the vulnerability, by virtue of increased infrastructure and human habitation and dependence on vulnerable HEP schemes. There are ways of measuring both, as will be discussed later.

The Palcacocha fiasco

On the 8^th April 2003 NASA published a press release (Steitz and Buis, 2003) in relation to its ASTER satellite imagery that led to headlines such as Glacier crack places Peruvian city in peril (New Scientist) and then went on to say that A glacial flood is threatening to sweep away a Peruvian city, satellite imagery has revealed. The timing of this could not have been worse as the media in Peru picked this up and splashed it across the headlines just before the Easter vacation time when many people from Lima traditionally go to Huaraz for their Easter break. The local tourism industry collapsed almost overnight. Travel companies had mass cancellations of coach and bus journeys and hotels and hostels also lost many reservations. In the back of people’s minds were the catastrophes in Huaraz in 1941 when 6,000 people lost their lives as a result of an aluviòn, and other similar events within the Cordillera Blanca (Reynolds, 1992). The NASA team had identified what they thought was a massive crack in the glaciers below Palcaraju and Pucaranra peaks and below which meant the glacier tongue could collapse into the glacial lake leading to a huge displacement wave that would inundate Huaraz, as it had done in 1941. This misinterpretation of the ASTER satellite image was probably based on two distinct linear features, which are in fact only boundaries of different morphological units of the glacier amphitheatre (Vilīmek et al., 2005). There was no danger of the glacier collapsing as inferred by NASA. The economic losses of this non-physical disaster were estimated by the local mayor to be in excess of $20 million. Despite pleas to remove the offending press release from NASA’s website it still remains available. New Scientist also refused to publish a rebuttal stating that is was good for circulation figures. I should say that ASTER imagery is very good but it needs to be interpreted by people experienced in the terrain imaged.

When ICIMOD and UNEP published their glacier inventory for Nepal and Bhutan there was a great media fanfare. There were many quotes of the numbers of glacial lakes imminently about to burst and it certainly raised awareness although it tended to sensationalise the situation. A list of glacial lakes thought to be dangerous was published but little thought had been given to the criteria by which such a statement could be justified. Similarly, Imja Tsho is often stated as being dangerous because it was cited in the infamous list within the UNEP/ICIMOD inventory. Yet two seasons of field work at Imja Tsho, specifically to investigate the glacial hazards there, indicate that, while it should be monitored regularly, it poses no immediate hazard. How do we know? I will explain in the next section.

The media will always be prone to over-dramatisation and sensationalism and those interviewed by the media will always run the risk of being misquoted or having words put in our mouths or having statements selectively edited until quite a different meaning emerges. That’s the media! Not talking to the media is not the solution. We need to educate the media and have material available in a form that is readily available to the general public. By stating things ourselves in a more measured and qualified way, perhaps might have some benefit and balance the extremes.

THE PRESENT SITUATION AND THE WAY FORWARD

By 2000, it was clear that, while glacial hazards had been accepted as recognisable issue by the World Bank and UN Agencies, people were still describing glacial lakes as being dangerous without any set of agreed criteria. The responses were emotive and subjective. In order to address this, RGSL was awarded a major 3-year contract (2000-2003) by the Department for International Development, British Government, to develop glacial hazard and risk minimisation protocols in rural environments. We were able to bring together experience and expertise gained through many projects in Nepal, Bhutan and Peru to try to identify and test selected criteria. The output of this project was the first set of international criteria by which to perform glacial hazard assessment and is available online either through RGSL’s web-site (www.geologyuk.com) or through the British Geological Survey’s web-site (www.bgs.ac.uk; DFID Knowledge and Research portal, then Search for ‘Glacial hazards’). This has subsequently been adopted by the then International Commission for Snow and Ice (ICSI*) Working Group on Glacial and Permafrost Hazards. (*ICSI is now the Union Commission for the Cryospheric Sciences).

The protocols describe criteria that can be measured and how to measure them so there is now no excuse for any organisation not to know what is involved in undertaking a glacial hazard assessment. Furthermore, the techniques lead to obtaining objective measures. This approach was developed through work in Bhutan to form a Multi-Criteria Analysis (Dyce and Reynolds, 2002) whereby different parameters could be determined and weighted and then used to produce a GLOF Hazard Score for want of a better term. This approach was further developed dividing the parameters so as to be able to derive a Trigger Potential Index and also a Threshold Parameter Index. When cross-plotted against a given scale these give a clear indication of the relative hazard of a particular glacial lake. All the criteria can be measured objectively by relatively inexperienced staff to produce an indication of the overall Glacial Hazard rating. We have applied this in 2004 across an entire catchment of around 2,500 km² in Southern Tibet/China and been able to produce a single graphic with all the significant glacial lakes plotted according to their Trigger Potential and Threshold Parameter indices and hence prioritised. It also provides managers with a means of determining the relative benefits of different approaches to reduce the overall hazard – reducing the threshold or managing the trigger potential, for example. The MCA approach can be developed further to incorporate vulnerability map information as part of a larger Risk Management Strategy.

Furthermore, we were also concerned (Reynolds and Taylor, 2004), about the quality and reliability of the UNEP/ICIMOD inventory reports (Mool et al., 2001a and b) and that remote sensing techniques had been under-utilised and that much more could be achieved using such data. We were fortunate to be able to part-fund and collaborate with the Centre for Glaciology at the University of Wales, Aberystwyth, U.K., on a Knowledge Transfer Partnership project for three years to review and develop remote sensing methods for glacial hazard assessment. This has led to a number of important but unique techniques being developed (Quincey et al., 2005).

Again to cut a long story short, it is now possible to identify and map glacial lakes using remote sensing techniques and to produce Digital Elevation Models from stereo satellite images; to derive an inventory of glaciers and map all glacial lakes using both manual and semi-automatic land classification procedures; to monitor flow rates as small as 2 cm/day for debris-covered glaciers using Synthetic Aperture Radar imagery; and to map where proto-supra-glacial lakes are most likely to develop in the next two to three decades. Working with colleagues originally at the University of Zurich it is possible to calculate and map the probability of inundation from debris flows and glacial lake outburst floods. Since 1996 we have also developed and tested different geophysical techniques on moraines to determine if they are ice-cored or not at a wide variety of Himalayan glacial lakes (e.g. Delisle et al., 2003; Hanisch et al., 1998; Pant and Reynolds, 2000; Reynolds, 2006).

It is my long held view that glacial hazard assessment and remediation should be part of an integrated rural development strategy that looks at wider issues including socio-economic factors in relation to risk management. It is no longer acceptable to view glacial lakes as a single-issue without recourse to examination of the wider environs. Emotive and subjective views are not acceptable and should be exposed for what they are.

The technology, which I have touched upon briefly, is now available to provide the tools with which to determine glacial hazards more robustly and objectively. It is now down to the political will and financial resources being available to harness and utilise these tools for the benefit of mountain communities. Cynically I fear that such resources will not be forthcoming until after another large disaster such is the complacency of the donor community.

CONCLUSIONS

Glacial hazards are increasing with climate change and we can expect more GLOFs to occur in the future and perhaps with greater vigour. Many people, communities and infrastructure are vulnerable. However, there are now clear objective methods available, using both remote sensing and ground-based investigation techniques to populate internationally-recognised protocols for glacial hazard assessment. It is now possible to derive relative Glacial Hazard ratings for glacial lakes using objective criteria. What is needed is a clear unequivocal assessment of the scale of the problem not only throughout Nepal but across the Himalayas. The tools are now available, the imagery could be obtained, and the rapid methods of analysing the imagery exist.

Dependence upon incomplete, isolated and fragmented data can only lead to a misrepresentation of the situation regarding the glaciers of the Himalayas. There is still much to learn about the linkage between glacier response and climate change, yet despite the paucity of information people have been making unjustified extrapolations about the possible disappearance of glaciers in the Himalayas. Much depends of having accurate information. Not only are the hazard assessments important but so too are the implications for water resource management and all that is consequent upon that.

There is still much to be done at Tsho Rolpa and for the people of Rolwaling and downstream in relation to glacial hazards within the region. It is looking hopeful that more funding might be forthcoming for a further assessment of the glacial hazard situation at Tsho Rolpa within the next two years and for a remote sensing monitoring programme for Rolwaling and the catchment to the north (draining into the Tama Kosi). This is but one glacial lake amongst a number that has not adequately been defined. One fact is clear – with time, other glaciers in Nepal will develop glacial lakes like that at Tsho Rolpa, only significantly larger. But we have the wherewithal to monitor the situation and consider the possible consequences objectively and calmly and to work with the relevant authorities if the resources are made available. If no progress is made, the information vacuum will only serve to fuel the doom and gloom merchants who will continue to use out-of-date or inappropriate information to make extrapolations that are spurious, misleading and potentially dangerous. On the other hand, saying that nothing much will happen will only give the authorities and donor community the excuse they need to continue to do nothing.

Postscript

This article is but the tip of an iceberg and there is much more behind what I have written here that time and space preclude me from expanding on. If there are omissions in the completeness of my recollections, I apologise, unless they are deliberate. If readers want more technical details relating to some of the issues I have described please refer to RGSL’s Web site under Glacial Hazards.

References

BPC Hydroconsult. 1997. Tsho Ropa GLOF Risk Reduction Project Formulation Mission Final Report. December 1997. BPC Hydroconsult, Kathmandu, Nepal.

Damen, M. 1992. Study of potential outburst flooding of Tsho Rolpa glacier lake, Rolwaling Valley, East Nepal. International Institute for Aerospace Survey and Earth Sciences, ITC, Enschede, The Netherlands.

Delisle, G., Reynolds, J.M., Hanisch, J., Pokhrel, A.P., & Pant, S. 2003. Lake Thulagi/Nepal: rapid landscape evolution in reaction to climate change. Annals of Geomorphology, 130:1-9.

Hanisch, J., Delisle, G., Pokhrel, A.P., Dixit, A.M., Reynolds, J.M., & Grabs, W.E. 1998. The Thulagi glacier lake, Manaslu Himal, Nepal - Hazard assessment of a potential outburst. Proceedings of the Eighth International IAEG Congress, 21-25 September 1998, Vancouver, Canada, pp. 2209-2215.

Modder, S. and Van Olden, Q. 1995. Geotechnical hazard analysis of a natural moraine dam in Nepal. June 1995 (Free University of Amsterdam).

Mool, P.K. et al. 2001a. Inventory of glaciers, glacial lakes and glacial lake outburst floods, Nepal. ICIMOD, Kathmandu, and UNEP, Nairobi.

Mool, P.K. et al. 2001b. Inventory of glaciers, glacial lakes and glacial lake outburst floods, Bhutan. ICIMOD, Kathmandu, and UNEP, Nairobi.

OYO Corporation. 1995. Electrical resistivity exploration at Tsho Rolpa end moraine. OYO Corporation, Tokyo, Japan (February 1995).

Pant, S.R. and Reynolds, J.M. 2000. Application of electrical imaging techniques for the investigation of natural dams: an example from the Thulagi Glacier Lake, Nepal. Journal of the Nepal Geological Society, 22:211-218.

Paterson, W.S.B. 1994. The physics of glaciers. Pergamon. Third edition, 480 pp.

Quincey, D.J., Lucas, R.M., Richardson, S.D., Glasser, N.F., Hambrey, M.J., and Reynolds, J.M. 2005. Optical remote sensing techniques in high-mountain environments: application to glacial hazards. Progress in Physical Geography, 29(4):475-505.

Rana, B., Shrestha, A.B., Reynolds, J.M., Aryal, R., Pokhrel, A.P. and Budhathoki, K.P. 2000. Hazard assessment of the Tsho Rolpa Glacier Lake and ongoing remediation measures. Journal of the Nepal Geological Society, 22:563-570.

Reynolds Geo-Sciences Ltd. 1994. Hazard assessment at Tsho Rolpa, Rolwaling Himal, northern Nepal: technical report. Report No: J9402.002 (December 1994), RGSL, Mold, UK.

Reynolds Geo-Sciences Ltd. 1996. Final proposal for the lowering of the water level at Tsho Rolpa Glacier Lake, Nepal, to minimise potential GLOF hazards. Report No: J9513.009 (February 1996), RGSL, Mold, UK.

Reynolds Geo-Sciences Ltd. 1997a. Assessment of the present status of the Tsho Rolpa Glacier Lake in Dolakha District of Nepal. Report No: J9735.024 (June 1997), RGSL, Mold, UK.

Reynolds Geo-Sciences Ltd. 1997b. Assessment of the present status of the Tsho Rolpa Glacier Lake in Dolakha District of Nepal – Supplementary Report. Report No: J9735.027 (June 1997), RGSL, Mold, UK.

Reynolds Geo-Sciences Ltd. 1999. Structural glaciological analysis of Trakarding Glacier, Rolwaling Himal, Nepal. Report No: J9857.049 (April 1999), RGSL, Mold, UK.

Reynolds, J.M. 1998. High altitude glacial lake hazard assessment and mitigation: a Himalayan perspective. In: Maund, J. & Eddleston, M. (eds.) Geohazards in engineering geology. Geological Society Engineering Group Special Publication No. 15, pp. 25-34.

Reynolds, J.M. 1992 The identification and mitigation of glacier-related hazards: examples from the Cordillera Blanca, Peru. In: McCall, G.J.H., Laming, D.C.J. and Scott, S. (eds), Geohazards, London, Chapman & Hall, pp. 143-157.

Reynolds, J.M. 1998. Managing the risks of glacial flooding at hydro plants. Hydro Review Worldwide, 6(2):18-22.

Reynolds, J.M. 1999. Photographic feature: Glacial hazard assessment at Tsho Rolpa, Rolwaling, Central Nepal. Quarterly Journal of Engineering Geology, 32(3):209-214.

Reynolds, J.M. 2006. The role of geophysics in glacial hazard assessment. First Break 24(9):61-66.

Reynolds, J.M. and Taylor, P.J. 2004. Review of ‘Inventory of glaciers, glacial lakes and Glacial Lake Outburst Floods, monitoring and Early Warning Systems in the Hindu Kush-Himalaya Region: Nepal and Bhutan (2 volumes)’, Mountain Research and Development, 24(3):272-274.

Richardson, S.D. and Reynolds, J.M. 2000. An overview of glacial hazards in the Himalayas. Quaternary International, 65/66(1):31-47.

Steitz, D.E. and Buis, A. 2003. NASA Press release 03-138. http//www.nasa.gov/home/ hqnews/2003/apr/HP_News_03138.html.

Vilīmek, V., Zapata, M.L., Klimeš, J., Patzelt, Z. and Santillán, N. 2005. Influence of glacial retreat on natural hazards on the Palcacocha Lake area, Peru. Landslides, 2(2):107-115.

WECS. 1993. Interim report on the field investigation on the Tsho Rolpa Glacier Lake, Rolwaling Valley. WECS Repoirt No. 3/4/021193/1/1 Seq. 436, Water and Energy Commission Secretariat, Kathmandu, Nepal.

Yamada, T. 1996. Report of investigations of Tsho Rolpa Glacier Lake, Rolwaling Valley. WECS/ JICA, August 1996.