Glacier Behaviour Research Articles

Drainage and refill of an Antarctic Peninsula subglacial lake reveal an active subglacial hydrological network

Abstract. The presence of subglacial lakes and subglacial hydrological networks under the East and West Antarctic ice sheets is now relatively well understood, whilst their influence on ice dynamics is the subject of ongoing research. In contrast, little is known about subglacial lakes and hydrological networks under the Antarctic Peninsula Ice Sheet and how these are influencing glacier behaviour. Here we describe the rapid drainage and slow refill of a subglacial lake under Mars Glacier using remote sensing and aerogeophysics. Results suggest drainage of the subglacial lake occurred prior to 2011, resulting in the collapse of the overlying ice into the newly formed subglacial cavity. The cavity has been refilling since this time, with peak rates of infilling associated with seasonal surface meltwater activity. We review evidence for similar features elsewhere in the Antarctic Peninsula and discuss whether their appearance marks a threshold shift in glacier thermal regimes and the activation or enhancement of their subglacial hydrological networks by surface meltwater. Collectively, these features show coupling of climate processes and the bed of the region's glaciers highlighting their ongoing vulnerability to climate change.

Glacial history and depositional environments in little Storfjorden and Hambergbukta of Arctic Svalbard since the younger dryas

Geophysical and lithological data provide crucial information for the understanding of glacial history in Arctic Svalbard. In this study, we reconstructed the glacier-induced depositional environments of Little Storfjorden and its tributary, Hambergbukta, over the last 13 ka to better understand the glacial history of southeastern Svalbard. The combined uses of swath-bathymetry, high-resolution seismic stratigraphy, and multiple-proxy measurements of sediment cores allowed us to define five steps of glacier-induced depositional environments: 1) deposition of massive, semi-consolidated gravelly sandy mud (Facies 1) during re-advance or still-stand of the marine-based glaciers/ice streams in Little Storfjorden during Younger Dryas (13–12 ka); 2) deposition of massive mud to gravelly sandy mud (Facies 2A and B) during glacial retreat until the earliest Holocene (12–10.1 ka); 3) sediment winnowing by enhanced bottom currents during the early to middle Holocene (10.1–3.7 ka); 4) deposition of bioturbated sandy mud (Facies 3) with high productivity under seasonal sea ice conditions during the late Holocene (3.7–0.7 ka); and 5) deposition of (slightly) bioturbated sandy to gravelly mud (Facies 4) affected by glacier surges since Little Ice Age (LIA) (Facies 4). In addition to seismic stratigraphy, depositional patterns of IRD in Little Storfjorden indicate that the glacier surges in Hambergbukta occurred only after ∼0.7 ka. This suggests that the terminal moraine complex (TMC) represents the maximum extent of the LIA surges, which argues against the recent inference for the TMC formation during pre-LIA. This study shows the importance of multiple parameters to better understand the current behavior of tidewater glaciers in the Svalbard fjords in response to rapid climate change.

Spatiotemporal variations in surface albedo during the ablation season and linkages with the annual mass balance on Muz Taw Glacier, Altai Mountains

ABSTRACT Melt-albedo feedback on glaciers is recognized as important processes for understanding glacier behavior and its sensitivity to climate change. This study selected the Muz Taw Glacier in the Altai Mountains to investigate the spatiotemporal variations in albedo and their linkages with mass balance, which will improve our knowledge of the recent acceleration of regional glacier shrinkage. Based on the Landsat-derived albedo, the spatial distribution of ablation-period albedo was characterized by a general increase with elevation, and significant east–west differences at the same elevation. The gap-filling MODIS values captured a nonsignificant negative trend of mean ablation-period albedo since 2000, with a total decrease of approximately 4.2%. From May to September, glacier-wide albedo exhibited pronounced V-shaped seasonal variability. A significant decrease in annual minimum albedo was found from 2000 to 2021, with the rate of approximately −0.30% yr−1 at the 99% confidence level. The bivariate relationship demonstrated that the change of ablation-period albedo explained 82% of the annual mass-balance variability. We applied the albedo method to estimate annual mass balance over the period 2000–2015. Combined with observed values, the average mass balance was −0.82 ± 0.32 m w.e. yr−1 between 2000 and 2020, with accelerated mass loss.

Past and future behavior of the valley glaciers in the Italian Alps

The behavior of the valley glaciers of the Italian Alps as a result of the climate changes expected for the 21st century has been investigated. From 1980 to 2017 the average length reduction of these glaciers has been 13% and their average areal reduction was around 22%. The mean climate sensitivity of the considered glaciers is 410 mK−1 and their mean observed shortening was about 500 m for a temperature increase of 1.4°C. To quantify the valley glacier life expectancy, a model estimating their length variations from the air temperature variations of the EuroCordex climatological projections of six different models under the RCP4.5 and RCP8.5 scenarios has been used. The ensemble mean temperatures in the Italian Alps region under these scenarios indicate increases in the temperature of ∼2°C and ∼4°C from 2018 to 2100 respectively. The glacier projections are mean values, taking into account the errors due to the model (6–16%) and the different climatological forcings (up to 20–30%). Under the RCP4.5 scenario, the glacier model projections show a constant retreat until the 2080s, with a weakening around 2090s and the 87% of glaciers surviving. Under the RCP8.5 scenario, only three glaciers could reach the end of the century, while 80% of them will probably disappear between the 2060s and the 2080s.

Simulation of the potential impacts of lakes on glacier behavior over the Tibetan Plateau in summer

Simulation of the potential impacts of lakes on glacier behavior over the Tibetan Plateau in summer

Influence of Particulate Matter on the Albedo of Qiangtang No. 1 Glacier, Tibetan Plateau

The melting behavior of glaciers on and around the Tibetan Plateau is strongly influenced by their albedo. In this paper, we report continuous observations made on the Qiangtang (QT) No. 1 Glacier, located in the central Tibetan Plateau, during its 2013–2015 melting seasons. Surface snow on the QT No. 1 Glacier mainly had a dust content less than 600 ppm and a black carbon (BC) content less than 10 ppb. A strong negative correlation was observed between albedo and dust content up to a threshold concentration of 1000 ppm, although albedo remained constant when dust concentrations increased above this value. The radii of snow particles showed a log-normal distribution that had a mean value of ~500 μm, but maximum and minimum values of 2539 μm and 40 μm, respectively. Snow density showed a normal distribution with a total range of 193–555 kg/m3, although most snow had a density of 400 kg/m3. Snow, ice, and aerosol radiative (SNICAR) simulations showed that dust and BC in the surface snow of the QT No. 1 Glacier reduced the snow and ice albedo by 5.9% and 0.06%, respectively, during the ablation season in 2015; however, the simulated particle impact was greater than the albedo reduction measured from field data. We interpret that dust has played a significantly more important role in melting of the QT No. 1 Glacier than BC over the study period, which is mainly due to the scarcity of human activities in the region and the low concentration of BC being produced.

Hydrometeorological progression of the Himalayan cryosphere: The Karakoram predicament

• Formulation of AI for glaciers using snowfall, snowmelt and temperature trends. • AI-based sorting of glacier-negative (losing mass) and glacier-positive (gaining mass). • Meteorological, stable isotope and geomorphic proof for glacier-positive behaviour in Karakoram. • Investigation of glacier-positive behaviour outside of the Karakoram region. • Linking AI to Himalayan precipitation periodicity via IOD, ISM and MLW connectivity. The anomalous behaviour of the Karakoram glaciers has been a decade-long discussion. While a lot of effort has been given to understanding this phenomenon, little has been explored with respect to the factors that govern glacier accumulation rates. We performed a time-series analysis of 17 Himalayan glaciers to evaluate the Accumulation Index (AI), examine the ice accumulation/ablation balance of each glacier, and investigate the natural periodicities in regional climatic behaviour. Based on the AI, the Himalayan glaciers have been characterised as ‘glacier-positive’ and ‘glacier-negative’. Glacier-positive behaviour in the Karakoram is favoured by global meteorological and local topographical factors. Analysis of 120 years of meteorological data shows a significant positive trend in winter and annual precipitation in the northwestern Himalayas, linked to surfacial warming and the Indian Ocean Dipole, while local factors such as slope, aspect and supraglacial debris constrain summer ablation. Additionally, the Karakoram precipitation is periodic in nature, demonstrating anti-correlation with decadal-scale precipitation records from the remaining Himalayan glaciers. Both local and global factors explain the anomaly in the Karakoram or the lack of it in other Himalayan glaciers. In a potentially warming planet, glacial accumulation rates are becoming increasingly sensitive to ambient temperatures, where glacier mass budget turns sharply negative at temperatures above 263 K due to accelerated summer ablation which overcompensates wintertime accumulation. This makes the Karakoram more sensitive to global meteorology than local factors. Therefore, it is important to address whether the Karakoram anomaly is an isolated climatic event or will it succumb to global climate change like the remainder of the Himalayan glaciers.

Analysis of differential glacier behaviour in Sikkim Himalayas in view of changing climate

The assessment of temporal changes in glacier response due to climatic change is critical from the hydrological perspective. The present study aims to identify temporal changes in glacier response using area changes, retreat, and surface elevation changes. The Friedman test and Wilcoxon signed-rank test were introduced to identify the temporal changes in the parameters mentioned above for the entire Sikkim from a sample of 28 studied glaciers. The result shows the area changes lies between 0.044–0.462, 0.143–0.351, 0.083–0.685, and 0.157–1.39%y−1 between 1988–2000, 2000–08, 2008–14 and 2014–18, respectively. Further, the magnitude of deglaciation has increased after 2014 by 73%, 76%, and 100% compared to the third, second, and first timeframe. Likewise, retreat is 10.82 ± 3.53, 19.07 ± 4.19,13.32 ± 5.59,23.44 ± 5.3 my−1 between 1998–2000, 2000–08, 2008–14 and 2014–18, respectively in the sample glaciers. Additionally, surface elevation change shows a negative-accelerated trend in the sample glaciers from 0.15 to 1.92 my−1 to 0.88 to 2.52 my−1 between 2000–08 and 2008–18, respectively.

The Last Glaciation in Valchiavenna (Italian Alps): maximum ice elevation data and recessional glacial deposits and landforms

This work presents the first extensive, 1:10,000-scale field survey data concerning glacial deposits and glacigenic landforms in the Valchiavenna territory, which has an area of 578 km2. Valchiavenna is an inner Alpine valley in Northern Italy between the Lepontine and Western Rhaetian Alps.A comprehensive 1:25,000 map of deposits and landforms from the last glaciation to the present is provided, describing i) glacial trimline evidence and associated features, such as moraine ridges, erratic boulders, ice- moulded bedrock surfaces and kame terraces; ii) glacial, ice-contact, lacustrine/peat and gravity-reworked till deposits; iii) other supraglacial, marginoglacial and subglacial landforms; and iv) erratics in glacial deposits. Establishing an absolute chronology of glacier dynamics was not the objective of this work. However, a relative chronology was inferred from sedimentological and geomorphological evidence: this allowed the description of the general behaviour of glaciers in the area during and after the Last Glacial Maximum (LGM).The palaeogeography at the LGM and the palaeo-ice-flow pattern were reconstructed on the basis of field data; this data confirmed that the valleys were almost completely filled by glacier ice, covering about 88% of the study area, with only the most elevated ridges and a few nunataks emerging above the ice surface, and allowed the identification of different source areas for the erratics found on opposite sides of the main valley.The observation of stratigraphical and geomorphological relationships between glacial deposits and landforms made it possible to propose a relative chronology of glacial advances and to outline the general glacial dynamics of the area. Both at the LGM and during the deglaciation after the LGM, the valley glacier inserted offshoots in tributary valleys, thus generally blocking the advance of local glaciers. With the gradual melting of the valley glacier during the deglaciation after the LGM, tributary glaciers could deposit tills on areas previously covered by valley glacier ice and at lower altitudes than the older lateral moraines. The main outcome of this work is a rich and homogeneous database of glacial deposits and glacigenic landforms that will be useful for further local and regional studies. It can guide the planning of geochronological dating and represents a fundamental step in the identification of glacial stadials and ice mass modelling. It can also support biogeography studies and the evaluation of the effects of climate change, slope dynamics modelling and hazard prediction.

Recent contrasting behaviour of mountain glaciers across the European High Arctic revealed by ArcticDEM data

Abstract. Small land-terminating mountain glaciers are a widespread and important element of Arctic ecosystems, influencing local hydrology, microclimate, and ecology. Due to their relatively small ice volumes, this class of ice mass is particularly sensitive to the significant ongoing climate warming in the European sector of the Arctic, i.e. in the Barents Sea area. Archipelagos surrounding the Barents Sea, i.e. Svalbard (SV), Novaya Zemlya (NZ), and Franz Josef Land (FJ), host numerous populations of mountain glaciers, but their response to recent strong warming remains understudied in most locations. This paper aims to obtain a snapshot of their state by utilizing high-resolution elevation data (ArcticDEM) to investigate the recent (ca. 2011–2017) elevation and volume changes of 382 small glaciers across SV, NZ, and FJ. The study concludes that many mountain glacier sites across the Barents Sea have been in a critical imbalance with the recent climate and might melt away within the coming several decades. However, deviations from the general trend exist; e.g. a cluster of small glaciers in north SV has been experiencing thickening. The findings reveal that near-stagnant glaciers might exhibit contrasting behaviours (fast thinning vs. thickening) over relatively short distances, which is a challenge for glacier mass balance models but also an opportunity to test their reliability.

Cosmogenic 10Be constraints on deglacial snowline rise in the Southern Alps, New Zealand

Geochronological dating of glacial landforms, such as terminal and lateral moraines, is useful for determining the extent and timing of past glaciation and for reconstructing the magnitude and rate of past climate changes. In the Southern Alps of New Zealand, well-dated glacial geomorphological records constrain the last glacial cycle across much of the Waitaki River basin (e.g. Ōhau, Pukaki, Tekapo) but its southern sector such as the Ahuriri River valley remains comparatively unconstrained. Recently, there has been debate on the scale and rapidity of mountain glacier retreat during the last glacial termination, particularly the 20–17 ka period in New Zealand. Missing from this debate is well-constrained equilibrium-line altitude (ELA) and associated temperature reconstructions, particularly over the period around 17 ka, which can help us to develop a more complete picture of how past temperature changes drove glacier retreat. Here we report the first glacial chronology dataset from the Last Glacial Maximum (LGM) and subsequent deglaciation from the Ahuriri River valley, Southern Alps, New Zealand (44°23′54″S, 169°39′48″E) based on 38 beryllium-10 (10Be) surface-exposure ages from terminal moraine systems and glaciated bedrock situated at the lower and middle sections of the valley. Our results show that the former Ahuriri Glacier reached its maximum extent at 19.8 ± 0.3 ka, which coincides with the global Last Glacial Maximum. By 16.7 ± 0.3 ka, the glacier had retreat ∼18 km up-valley suggesting at least ∼43% glacier-length loss relative to its full LGM extent. This deglaciation was accompanied by the formation of a shallow proglacial lake. Using the accumulation area ratio (AAR) method, we estimate that the ELA was lower than present by ∼880 m (∼1120 m a.s.l.) at 19.8 ± 0.3 ka, and ∼770 m lower (∼1230 m a.s.l.) at 16.7 ± 0.3 ka. Applying an estimate for temperature lapse rate, this ELA anomaly implies that local air temperature was 5 ± 1 °C colder than present (1981–2010) at 19.8 ± 0.3 ka, while it was 4.4 ± 0.9 °C colder at 16.7 ± 0.3 ka, assuming no change in precipitation. The substantial glacier retreat in response to a relatively small accompanying increases in ELA (110 m) and temperature (0.6 °C) may have been a result of the high glacier-length sensitivity of this glacier system due to its low gradient of former ice surface. Our low warming estimate differs markedly from other deglaciation studies, specifically from Rakaia River valley, which reports a much larger temperature increase at the onset of the last deglaciation. This precisely-dated moraine record along with reconstructed ELA as proxies for atmospheric conditions, provides new insight into post LGM glacier behaviour and climate conditions in New Zealand.

Assessing the Contribution of Glacier Melt to Discharge in the Tropics: The Case of Study of the Antisana Glacier 12 in Ecuador

Tropical glaciers are excellent indicators of climate variability due to their fast response to temperature and precipitation variations. At same time, they supply freshwater to downstream populations. In this study, a hydro-glaciological model was adapted to analyze the influence of meteorological forcing on melting and discharge variations at Glacier 12 of Antisana volcano (4,735–5,720 m above sea level (a.s.l.), 1.68 km2, 0°29′S; 78°9′W). Energy fluxes and melting were calculated using a distributed surface energy balance model using 20 altitude bands from glacier snout to the summit at 30-min resolution for 684 days between 2011 and 2013. The discharge was computed using linear reservoirs for snow, firn, ice, and moraine zones. Meteorological variables were recorded at 4,750 m.a.s.l. in the ablation area and distributed through the altitudinal range using geometrical corrections, and measured lapse rate. The annual specific mass balance (−0.61 m of water equivalent -m w.e. y−1-) and the ablation gradient (22.76 kg m−2 m−1) agree with the values estimated from direct measurements. Sequential validations allowed the simulated discharge to reproduce hourly and daily discharge variability at the outlet of the catchment. The latter confirmed discharge simulated (0.187 m3 s−1) overestimates the streamflow measured. Hence it did not reflect the net meltwater production due to possible losses through the complex geology of the site. The lack of seasonality in cloud cover and incident short-wave radiation force the reflected short-wave radiation via albedo to drive melting energy from January to June and October to December. Whereas the wind speed was the most influencing variable during the July-September season. Results provide new insights on the behaviour of glaciers in the inner tropics since cloudiness and precipitation occur throughout the year yielding a constant short-wave attenuation and continuous variation of snow layer thickness.

The imprint of catchment processes on Greenlandic ice cap proglacial lake records: analytical approaches and palaeoenvironmental significance

ABSTRACTLakes fed by Greenlandic mountain glaciers and ice caps (GICs) contain important archives of Arctic palaeoenvironmental change. GIC proglacial lake records have been increasingly used to reconstruct Holocene glacier behaviour, largely focusing on macrostratigraphy. However, despite the wide range of topographic settings and catchment characteristics, there has been little systematic analysis of the ways that catchment conditions are registered in the clastic sediments of GIC lakes. Such signals provide valuable insights into landscape processes and palaeoenvironmental conditions that are not routinely captured in other Quaternary glacial morphosedimentary archives. This review synthesises sedimentological and geochemical evidence from existing Holocene GIC proglacial lake records to establish: how catchment‐wide conditions have been recorded in the lacustrine sequences; and our ability to isolate these signals to enhance palaeoenvironmental reconstruction. Our review shows that with careful sedimentological and targeted (bio)geochemical analyses coupled with a clear process‐based understanding, catchment and in‐lake signals can be effectively identified in the microstratigraphic and mineral grain record. Such signals include wind patterns, mass wasting, precipitation events and seasonal lake ice cover, that can complement broader palaeoclimatic proxy evidence. The approaches collated here, if more widely applied, could considerably enhance environmental reconstructions not only in Greenland, but in glaciated catchments elsewhere.

Documenting 20th and 21st century glacier change and landscape evolution with maps and land, aerial, and space-based geospatial imagery in Alaska’s Kenai Mountains

Data fusion and analysis of maps and remote sensing data collected from different spatial perspectives (ground, air, and space) at different times from the early 20th century to the present using different sensors were used to answer questions about glacier behavior and rapidly changing landscapes of Alaska’s southern Kenai Mountains. Expeditions to three fiords of the southern Kenai Mountains were conducted during the summers of 2004, 2005, 2006, and 2021. Each expedition used repeat photography to document glacier behavior and change, and landscape evolution at six Kenai Mountains glaciers, most located within Kenai Fjords National Park. Bear Glacier, Aialik Glacier, Pedersen Glacier, Holgate Glacier, Little Holgate Glacier, and Northwestern Glacier were studied and at a minimum, their terminus positions were determined for the following dates: 1909, 1950, 1961, 1973, 1990, 2004-2006, and 2021. Each glacier displayed unique asynchronous behavior. Since 1909, all displayed long-term terminus retreat. However, the timing for each glacier was unique. In 2021, Holgate Glacier was advancing, while the other five glaciers were retreating.

A regionally resolved inventory of High Mountain Asia surge-type glaciers, derived from a multi-factor remote sensing approach

Abstract. Knowledge about the occurrence and characteristics of surge-type glaciers is crucial due to the impact of surging on glacier melt and glacier-related hazards. One of the super-clusters of surge-type glaciers is High Mountain Asia (HMA). However, no consistent region-wide inventory of surge-type glaciers in HMA exists. We present a regionally resolved inventory of surge-type glaciers based on their behaviour across High Mountain Asia between 2000 and 2018. We identify surge-type behaviour from surface velocity, elevation and feature change patterns using a multi-factor remote sensing approach that combines yearly ITS_LIVE velocity data, DEM differences and very-high-resolution imagery (Bing Maps, Google Earth). Out of the ≈95 000 glaciers in HMA, we identified 666 that show diagnostic surge-type glacier behaviour between 2000 and 2018, which are mainly found in the Karakoram (223) and the Pamir regions (223). The total area covered by the 666 surge-type glaciers represents 19.5 % of the glacierized area in Randolph Glacier Inventory (RGI) V6.0 polygons in HMA. Only 68 glaciers were already identified as “surge type” in the RGI V6.0. We further validate 107 glaciers previously labelled as “probably surge type” and newly identify 491 glaciers, not previously reported in other inventories covering HMA. We finally discuss the possibility of self-organized criticality in glacier surges. Across all regions of HMA, the surge-affected area within glacier complexes displays a significant power law dependency with glacier length.

The Ticino-Toce glacier system (Swiss-Italian Alps) in the framework of the Alpine Last Glacial Maximum

We present a new glacier chronology from one of the major end moraine systems of the Southern Alps. Timing and extent of the Last Glacial Maximum (LGM) advance of the Ticino-Toce glacier were reconstructed in detail based on landform relationships and surface exposure dating. 10Be and 36Cl ages from 41 erratic boulders constrain the last maximum of the Ticino-Toce glacier to have lasted from 25.0 ± 0.9 ka to 19.9 ± 0.7 ka. Over this period of ca. 5000 years the Ticino-Toce glacier underwent only minor oscillations, during which the glacier front remained close to its LGM maximum position. Underpinned by our detailed geomorphological reconstruction, the ages further demonstrate that the Verbano piedmont lobe was significantly larger (about 200 km2) than shown on recent maps. A short-lived glacier readvance around 19.7 ± 1.1 ka terminated internal to the LGM maximum margins. Collapse of the Verbano lobe must have set in soon after and final withdrawal from the foreland occurred no later than 19 ka. The new Ticino-Toce chronology matches temporal constraints of major glacier systems emerging from adjacent Alpine accumulation areas. This points to in-phase glacier advances across the Western Alps, apart from the southwestern massifs for which a slightly earlier reaching of the Late Pleistocene maximum position of the Lyon lobe was shown recently. A two- potentially three-fold LGM advance structure with intervening retreat intervals, as reported for two major amphitheatres (Garda, Tagliamento) to the east is distinctly not supported by our data for the Ticino-Toce glacier. This contrast in glacier behaviour between the central and eastern Southern Alps is ascribed to shifting precipitation patterns after ca. 23–22 ka, away from southern to predominantly northwestern air flow and a greater independence of southerly moisture transport for the central Southern Alps.

Strong acceleration of glacier area loss in the Greater Caucasus between 2000 and 2020

Abstract. An updated glacier inventory is important for understanding glacier behaviour given the accelerating glacier retreat observed around the world. Here, we present data from a new glacier inventory for two points in time (2000, 2020) covering the entire Greater Caucasus (Georgia, Russia, and Azerbaijan). Satellite imagery (Landsat, Sentinel, SPOT) was used to conduct a remote-sensing survey of glacier change. The 30 m resolution Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM; 17 November 2011) was used to determine aspect, slope, and elevations, for all glaciers. Glacier margins were mapped manually and reveal that in 2000 the mountain range contained 2186 glaciers with a total glacier area of 1381.5 ± 58.2 km2. By 2020, the area had decreased to 1060.9 ± 33.6 km2 a reduction of 23.2 ± 3.8 % (320.6 ± 45.9 km2) or −1.16 % yr−1 over the last 20 years in the Greater Caucasus. Of the 2223 glaciers, 14 have an area > 10 km2, resulting in the 221.9 km2 or 20.9 % of total glacier area in 2020. The Bezengi Glacier with an area of 39.4 ± 0.9 km2 was the largest glacier mapped in the 2020 database. Glaciers between 1.0 and 5.0 km2 accounted for 478.1 km2 or 34.6 % in total area in 2000, while they accounted for 354.0 km2 or 33.4 % in total area in 2020. The rates of area shrinkage and mean elevation vary between the northern and southern and between the western, central, and eastern Greater Caucasus. Area shrinkage is significantly stronger in the eastern Greater Caucasus (−1.82 % yr−1), where most glaciers are very small. The observed increased summer temperatures and decreased winter precipitation along with increased Saharan dust deposition might be responsible for the predominantly negative mass balances of Djankuat and Garabashi glaciers with long-term measurements. Both glacier inventories are available from the Global Land Ice Measurements from Space (GLIMS) database and can be used for future studies.

Debris-covered glacier systems and associated glacial lake outburst flood hazards: challenges and prospects

Glaciers respond sensitively to climate variability and change, with associated impacts on meltwater production, sea-level rise and geomorphological hazards. There is a strong societal interest in understanding the current response of all types of glacier systems to climate change and how they will continue to evolve in the context of the whole glacierized landscape. In particular, understanding the current and future behaviour of debris-covered glaciers is a ‘hot topic’ in glaciological research because of concerns for water resources and glacier-related hazards. The state of these glaciers is closely related to various hazardous geomorphological processes which are relatively poorly understood. Understanding the implications of debris-covered glacier evolution requires a systems approach. This includes the interplay of various factors such as local geomorphology, ice ablation patterns, debris characteristics and glacier lake growth and development. Such a broader, contextualized understanding is prerequisite to identifying and monitoring the geohazards and hydrologic implications associated with changes in the debris-covered glacier system under future climate scenarios. This paper presents a comprehensive review of current knowledge of the debris-covered glacier landsystem. Specifically, we review state-of-the-art field-based and the remote sensing-based methods for monitoring debris-covered glacier characteristics and lakes and their evolution under future climate change. We advocate a holistic process-based framework for assessing hazards associated with moraine-dammed glacio-terminal lakes that are a projected end-member state for many debris-covered glaciers under a warming climate.

Nonlinear sensitivity of glacier mass balance to future climate change unveiled by deep learning

Glaciers and ice caps are experiencing strong mass losses worldwide, challenging water availability, hydropower generation, and ecosystems. Here, we perform the first-ever glacier evolution projections based on deep learning by modelling the 21st century glacier evolution in the French Alps. By the end of the century, we predict a glacier volume loss between 75 and 88%. Deep learning captures a nonlinear response of glaciers to air temperature and precipitation, improving the representation of extreme mass balance rates compared to linear statistical and temperature-index models. Our results confirm an over-sensitivity of temperature-index models, often used by large-scale studies, to future warming. We argue that such models can be suitable for steep mountain glaciers. However, glacier projections under low-emission scenarios and the behaviour of flatter glaciers and ice caps are likely to be biased by mass balance models with linear sensitivities, introducing long-term biases in sea-level rise and water resources projections.

De-icing landsystem model for the Universidad Glacier (34° S) in the Central Andes of Chile during the past ~660 years

Reconstructing latest Holocene (<1000 years) glacial landscape development in the Central Andes of Chile (30–35° S) is key for understanding the response of the cryosphere during periods of negative glacier mass balance, such as the current one. The excellently preserved glacial landscape produced during the latest ice advance and retreat cycle is of particular interest for examining the detailed response of glaciers to deglaciation. To establish a conceptual model of glacier behavior under warm and dry climatic conditions, we reconstructed and dated the recent glacial history of the Universidad Glacier (34° S) through detailed geomorphological mapping and 10Be cosmogenic surface exposure dating. Our mapping describes a landsystem that spans from the current ice front to ~3 km down-valley, where a mosaic of glacial landforms includes mounded relief; sinkholes; debris-filled stripes; moraine belts; flutings; and a prominent basal till plain. Our 10Be ages suggest that the Universidad Glacier has fluctuated in its forefield since the 13th – 15th centuries CE. We propose that the glacier evolved from a clean glacier to a debris-covered glacier, to an ice-cored moraine, and finally, to a massive dead-ice topography. This deglacial evolution intermittent and potentially reset by multiple standstills and/or re-advances during the overall retreat. The implication is that phases of active ice were followed by stagnation associated with progressive melting of dead ice under the supraglacial debris layer. Similar geomorphic features and processes are recorded in the present-day Universidad Glacier ablation zone, denoting a recurrent reconditioning over time analogous to the glacier's evolution during the latest Holocene.