Recent and future EVOlution of Glacial LAkes in China (EVOGLAC): Spatio-temporal diversity and hazard potential

Glacial lakes are expanding and forming in many mountainous regions of the world as a consequence of continuing if not accelerating glacier shrinkage (e.g., Bolch et al. 2012, Yao et al. 2012a, Gardner et al., 2013; Vaughan et al., 2013). In combination with altered stability of surrounding steep rock and ice walls, there is an increasing risk of Glacial Lake Outburst Floods (GLOFs) (Kaab et al., 2005b; Schaub et al., 2013; Clague and O’Connor, 2014; Deline et al., 2014). Nowhere is the threat more pronounced than in high mountain Asia, where residential, tourism, and particularly hydropower infrastructure is expanding higher into alpine valleys, increasing the potential for, and resulting in significant GLOF related disasters (Huggel et al., 2008). China comprises almost 50% of the Hindu Kush Himalayan land area alone (Sharma and Pratap, 1994), and contains considerably more potentially dangerous glacial lakes than seen across any of the other Himalayan countries (Ives et al., 2010; Wang et al., 2015a). Spanning large latitudinal and longitudinal gradients with diverse climate regimes ranging from humid monsoon to dry continental, the mountainous region of China is thus the ideal environment within which to further develop scientific understanding on lake formation processes and  build on recent advances in capabilities to anticipate (Allen et al., 2016b; Haeberli et al., 2016)and model (Westoby et al., 2014)GLOF threats. Ultimately this new knowledge will provide a robust and comprehensive scientific basis for GLOF hazard management and adaptation planning across high mountain Asia

Typically, GLOF disasters or perceived threats over recent decades have involved terminal or lateral moraine dams that formed during the Little Ice Age (between the years 1400 and 1900), and lakes that have filled during the subsequent thinning and retreat of glaciers during the twentieth century (Clague and O’Connor, 2014). While in general the threat from such lakes may be diminishing over time, assuming that the most unstable lakes would already have failed, the Himalaya are noted as one area where large proglacial lakes trapped behind Little Ice Age moraines are continuing to evolve (Clague and O’Connor, 2014). For such lake reservoirs developing on or at the margins of glaciers, remote sensing based methodologies and Geographic information Systems (GIS) have for more than a decade proven appropriate tools for monitoring hazardous developments across large spatial scales (e.g., Huggel et al., 2002; Wessels et al., 2002). While related large-scale (state, country to trans-national) inventories of GLOF hazard have since been implemented over much of high mountain Asia (e.g., Ives et al., 2010; Bolch et al., 2011; ICIMOD, 2011; Mergili and Schneider, 2011; Wang et al., 2011a; Tadono et al., 2012; Worni et al., 2013; Wang et al., 2015a), assessment schemes are heterogeneous and typically still require subjective case-by-case classification of potential triggering threats such as impacts from ice or rock avalanches.

In view of projected warming and continued retreat of alpine glaciers (Church et al., 2013), attention has shifted beyond monitoring and assessment of existing outburst threats, towards the anticipation of where new, potentially problematic lakes will form in the future (Haeberli et al., 2016). Such lakes are unlikely to be impounded by large moraine dams, which requires the glacier to remain stationary for a sufficient length of time, but will form in bedrock depressions or overdeepenings in the exposed glacier bed (Frey et al., 2010). While these lakes may form attractive landscape features, and even offer potential for hydropower generation (Haeberli and Hohmann, 2008), a primary concern is the potential threat from overtopping waves generated by mass movements of ice and rock, as warming may destabilize the surrounding steep slopes (Deline et al., 2014). Therefore, methods have recently been developed that enable not only the identification of where new lakes might form in the exposed bed topography (Linsbauer et al., 2012; Linsbauer et al., 2016), but also to recognize surrounding steep slopes from which mass movements may detach and impact into the glacial lakes (Schaub et al., 2013). However, a key limitation remains that any timing of the emergence and future evolution of glacial lakes in high mountain Asia is completely unconstrained. This represents a major scientific challenge, as  local geomorphological and climatological conditions will lead to significant diversity in lake evolution (Gardelle et al., 2011). In addition, integrated approaches are yet to be developed which consider other transient drivers of future GLOF hazard such as the altered stability of the surrounding rock and ice flanks, changes in hydrometeorlogical triggering (e.g. snow melt and rainfall), or thawing of the dam structures. 

The overall aim of the proposed study is to develop and implement a comprehensive methodological approach to investigate the recent and future evolution of glacial lakes and their related hazard potential in different climatic and geomorphological settings in China. This will lead to improved understanding and prediction of lake formation, change in GLOF triggering processes, and change in hazard in downstream areas, as glaciers continue to retreat over the 21^stcentury and beyond. The methodological approach will be suited for outscaling to larger regions, recognising the urgent need for robust scientific information to support adaptation planning in response to the rapidly evolving GLOF threat across high mountain Asia.

More specifically, the study will focus on the following research objectives:

·      Reconstruction of the spatio-temporal evolution of glacier thinning, retreat, and associated lake development since ca. 1970 in the three study regions with contrasting physical environmental settings.

·      Relation of the past evolution of lake development to local climatic, glaciological, geomorphic, and topographic characteristics, to derive key driving processes.

·      Based on relationships established above, catchment-scale modelling of future glacial retreat and lake development for given warming scenarios, considering both the expansion of existing lakes, but also the formation of new lakes in the exposed bed topography.

·      For given warming scenarios (and corresponding time-horizons), complete catchment-scale integrated modelling of GLOF potential, considering the number/size of lakes, and change in key triggering processes (ice-rock avalanches, snow-melt, rainfall, thawing of ice-cored moraine dams).

·      For selected critical current and future glacial lakes, complete GLOF modelling (coupled breach/overtopping and flow path simulation) and downstream hazard mapping. Conduct sensitivity testing for different lake volumes, triggering processes and dam characteristics to establish uncertainty in future scenarios.

·      Synthesis of the new understanding and preparation of a modelling chain for outscaling to other regions of high mountain Asia.

The study will focus on three distinct regions of Tibet with contrasting climatic regimes and physical environments (Figure 1):

1)    Boshula mountain range in Southeast Tibet (humid, monsoon-dominated climate)

2)    Poiqu river basin, Central Himalaya (humid subtropical climate, with significant precipitation gradient from south to north)

3)    Nyainqentanglha mountain range, Tibet Plateau (transition zone between monsoon influenced and continental climates)

The Chinese partner has long-standing experience conducting glaciology, glacial lake, and GLOF hazard studies in all three regions. The Swiss partner has previously worked in two of the study regions.

Figure 1: Overview of the Himalaya and Tibetan Plateau, indicating the location of the three proposed study regions. 

Boshula mountain range, southeast Tibet 

The Boshula mountain range is located at the boundary of Bomi and Paksho counties in the southeastern Tibetan Plateau. The elevation ranges from 3100 to 6200 m a.s.l and the topography is characterized by huge relief differences and steep slopes. A major national highway (Sichuan – Tibet) runs through this area. The region is positioned within the domain of the Indian monsoon, and hence is significantly influenced by warm and humid moisture in summer. A lake inventory from 2009 revealed 123 glacial lakes in the region, with an average expansion in lake area of 18.6% measured since the 1970s (Wang et al., 2011b). Up to 8 lakes have previously been considered as potentially dangerous (Wang et al., 2011a).

Poiqu river basin, Central Himalaya

The Poiqu River is a trans-boundary river that originates on the southern slopes of the Central Himalayas and flows southward into Nepal. The north-south orientated Poiqu River basin has a total area of 2018 km² within China, and elevation ranging from 1178 to 8012 m a.s.l. An important highway linking Nepal and China follows along the Poiqu River. Although under the influence of the Indian monsoon, there is a notable decrease in temperature and precipitation from south to north within the basin. A lake inventory from 2010 contained 119 lakes, with a total increase in lake area of 83.1% measured since the 1970s (Wang et al., 2015b). Up to 7 of these lakes were considered as potentially dangerous. 

Nyainqentanglha mountain range, Tibetan Plateau

The Nyainqentanglha mountain range, situated towards the southeast of the Tibetan Plateau is a southwest – northeast orientated mountain range extending for approximately 230 km in length with maximum elevations between 5000 and 7162 m a.s.l. The mountain range serves as a climate divide, with the catchments to the southeast influenced by the Indian Monsoon, while the catchments to the northwest are more characteristic of the dry continental climate of the central Tibetan Plateau. Recent changes in glacial extent have been mapped for this region (Bolch et al., 2010), but corresponding analyses of lake development are lacking.