Dr. Matthias Huss

The shrinkage of glaciers and the vanishing of glacier-fed streams (GFSs) are emblematic of climate change. However, forecasts of how GFS microbiome structure and function will change under projected climate change scenarios are lacking. Combining 2,333 prokaryotic metagenome-assembled genomes with climatic, glaciological, and environmental data collected by the Vanishing Glaciers project from 164 GFSs draining Earth’s major mountain ranges, we here predict the future of the GFS microbiome until the end of the century under various climate change scenarios. Our model framework is rooted in a space-for-time substitution design and leverages statistical learning approaches. We predict that declining environmental selection promotes primary production in GFSs, stimulating both bacterial biomass and biodiversity. Concomitantly, predictions suggest that the phylogenetic structure of the GFS microbiome will change and entire bacterial clades are at risk. Furthermore, genomic projections reveal that microbiome functions will shift, with intensified solar energy acquisition pathways, heterotrophy and algal-bacterial interactions. Altogether, we project a ‘greener’ future of the world’s GFSs accompanied by a loss of clades that have adapted to environmental harshness, with consequences for ecosystem functioning.

Glacier retreat presents significant environmental and social challenges. Understanding the local impacts of climatic drivers on glacier evolution is crucial, with mass balance being a central concept. This study introduces miniML-MB, a new minimal machine-learning model designed to estimate annual point surface mass balance (PMB) for very small datasets. Based on an eXtreme Gradient Boosting (XGBoost) architecture, miniML-MB is applied to model PMB at individual sites in the Swiss Alps, emphasising the need for an appropriate training framework and dimensionality reduction techniques. A substantial added value of miniML-MB is its data-driven identification of key climatic drivers of local mass balance. The best PMB prediction performance was achieved with two predictors: mean air temperature (May-August) and total precipitation (October-February). miniML-MB models PMB accurately from 1961 to 2021, with a mean absolute error (MAE) of 0.417 m w.e. across all sites. Notably, miniML-MB demonstrates similar and, in most cases, superior predictive capabilities compared to a simple positive degree-day (PDD) model (MAE of 0.541 m w.e.). Compared to the PDD model, miniML-MB is less effective at reproducing extreme mass balance values (e.g. 2022) that fall outside its training range. As such, miniML-MB shows promise as a gap-filling tool for sites with incomplete PMB measurements as long as the missing year's climate conditions are within the training range. This study underscores potential means for further refinement and broader applications of data-driven approaches in glaciology.

Glaciers are indicators of ongoing anthropogenic climate change¹. Their melting leads to increased local geohazards², and impacts marine³ and terrestrial^4,5 ecosystems, regional freshwater resources⁶, and both global water and energy cycles^7,8. Together with the Greenland and Antarctic ice sheets, glaciers are essential drivers of present^9,10 and future^{11, 12–13} sea-level rise. Previous assessments of global glacier mass changes have been hampered by spatial and temporal limitations and the heterogeneity of existing data series^{14, 15–16}. Here we show in an intercomparison exercise that glaciers worldwide lost 273 ± 16 gigatonnes in mass annually from 2000 to 2023, with an increase of 36 ± 10% from the first (2000–2011) to the second (2012–2023) half of the period. Since 2000, glaciers have lost between 2% and 39% of their ice regionally and about 5% globally. Glacier mass loss is about 18% larger than the loss from the Greenland Ice Sheet and more than twice that from the Antarctic Ice Sheet¹⁷. Our results arise from a scientific community effort to collect, homogenize, combine and analyse glacier mass changes from in situ and remote-sensing observations. Although our estimates are in agreement with findings from previous assessments^{14, 15–16} at a global scale, we found some large regional deviations owing to systematic differences among observation methods. Our results provide a refined baseline for better understanding observational differences and for calibrating model ensembles^12,16,18, which will help to narrow projection uncertainty for the twenty-first century^11,12,18.

We present forecasts of land-use/land-cover (LULC) change for Switzerland for three time-steps in the 21^st century under the representative concentration pathways 4.5 and 8.5, and at 100-m spatial and 14-class thematic resolution. We modelled the spatial suitability for each LULC class with a neural network (NN) using > 200 predictors and accounting for climate and policy changes. We improved model performance by using a data augmentation algorithm that synthetically increased the number of cells of underrepresented classes, resulting in an overall quantity disagreement of 0.053 and allocation disagreement of 0.15, which indicate good prediction accuracy. These class-specific spatial suitability maps outputted by the NN were then merged in a single LULC map per time-step using the CLUE-S algorithm, accounting for LULC demand for the future and a set of LULC transition rules. As the first LULC forecast for Switzerland at a thematic resolution comparable to available LULC maps for the past, this product lends itself to applications in land-use planning, resource management, ecological and hydraulic modelling, habitat restoration and conservation.

We present an extensive dataset of ice thickness measurements from Jostedalsbreen ice cap, mainland Europe's largest glacier. The dataset consists of more than 351 000 point values of ice thickness distributed along ∼1100 km profile segments that cover most of the ice cap. Ice thickness was measured during field campaigns in 2018, 2021, 2022 and 2023 using various ground-penetrating radar (GPR) systems with frequencies ranging between 2.5 and 500 MHz. A large majority of the ice thickness observations were collected in spring using either snowmobiles (90 %) or a helicopter-based radar system (8 %), while summer measurements were carried out on foot (2 %). To ensure accessibility and ease of use, metadata were attributed following the GlaThiDa dataset (GlaThiDa Consortium, 2020) and follow the FAIR (findable, accessible, interoperable and reusable) guiding principles. Our findings show that glacier ice of more than 400 m thickness is found in the upper regions of large outlet glaciers, with a maximum ice thickness of ∼630 m in the accumulation area of Tunsbergdalsbreen. Thin ice of less than 50 m covers narrow regions joining the central part of Jostedalsbreen with its northern and southern parts, making the ice cap vulnerable to break-up with future climate warming. Using the point values of ice thickness as input to an ice thickness model, we computed 10 m grids of ice thickness and bed topography that cover the entire ice cap. From these distributed datasets, we find that Jostedalsbreen (458 km² in 2019) has a present (∼2020) mean ice thickness of 154 ± 22 m and an ice volume of 70.6 ± 10.2 km3. Locations of depressions in the map of bed topography are used to delineate potential future lakes, consequently providing a glimpse of the landscape if the entire Jostedalsbreen melts away. Together, the comprehensive ice thickness point values and ice-cap-wide grids serve as a baseline for future climate change impact studies at Jostedalsbreen.

Even though approaches to artificially reduce local glacier melt have been developed, they face considerable challenges on the larger scale. To mitigate the negative effects of an imminent loss of mountain glaciers, preserving the ice by reducing greenhouse gas emissions remains the most effective solution.

Glacier mass balance reconstructions provide a means of placing relatively short observational records into a longer-term context. Here, we use multiple proxies from Pinus cembra trees from God da Tamangur, combining tree ring anatomy and stable isotope chronologies to reconstruct seasonal glacier mass balance (i.e., winter, summer, and annual mass balance) for the nearby Silvretta Glacier over the last 2 centuries. The combination of tree ring width, radial diameter of earlywood cell lumina, and latewood radial cell wall thickness provides a highly significant reconstruction for summer mass balance, whereas for the winter mass balance, the correlation was less significant but still robust when radial cell lumina were combined with δ¹⁸O records. A combination of the reconstructed winter and summer mass balances allows the quantification of the annual mass balance of the Silvretta Glacier for which in situ measurements date back to 1919. Our reconstruction indicates a substantial increase in glacier mass during the first half of the 19th century and an abrupt termination of this phase after the end of the Little Ice Age. Since the 1860s, negative glacier mass balances have been dominant and mass losses accelerate as anthropogenic warming picks up in the Alps.

Observations of glacier mass changes are key to understanding the response of glaciers to climate change and related impacts, such as regional runoff, ecosystem changes, and global sea level rise. Spaceborne optical and radar sensors make it possible to quantify glacier elevation changes, and thus multi-annual mass changes, on a regional and global scale. However, estimates from a growing number of studies show a wide range of results with differences often beyond uncertainty bounds. Here, we present the outcome of a community-based inter-comparison experiment using spaceborne optical stereo (ASTER) and synthetic aperture radar interferometry (TanDEM-X) data to estimate elevation changes for defined glaciers and target periods that pose different assessment challenges. Using provided or self-processed digital elevation models (DEMs) for five test sites, 12 research groups provided a total of 97 spaceborne elevation-change datasets using various processing approaches. Validation with airborne data showed that using an ensemble estimate is promising to reduce random errors from different instruments and processing methods but still requires a more comprehensive investigation and correction of systematic errors. We found that scene selection, DEM processing, and co-registration have the biggest impact on the results. Other processing steps, such as treating spatial data voids, differences in survey periods, or radar penetration, can still be important for individual cases. Future research should focus on testing different implementations of individual processing steps (e.g. co-registration) and addressing issues related to temporal corrections, radar penetration, glacier area changes, and density conversion. Finally, there is a clear need for our community to develop best practices, use open, reproducible software, and assess overall uncertainty to enhance inter-comparison and empower physical process insights across glacier elevation-change studies.

This study fills a gap in the long-term prediction of changes in parameters of the Elbrus glaciers, using the GloGEMflow-debris model to simulate the glacier evolution. The part 1 provides a detailed description of the model architecture. The model consists of three blocks in which the calculation of the surface mass balance, glacier flow and moraine transformation is carried out. The area and thickness of the moraine cover increase as glaciers degrade. This is important to consider, as a thicker layer of moraine reduces the ice melting. For predictive calculations, the data on temperature and precipitation for five SSP climate scenarios are taken from the CMIP6 project. A temperature index method is used to calculate the surface mass balance, taking into account the influence of the moraine cover: the ablation of pure ice is adjusted in accordance with the area and thickness of the moraine cover. The ice flow block is used to update the geometry of glaciers and moraine cover. The adaptation of the model to the glaciers of Elbrus includes the adjustment of the block of the moraine cover evolution, which corresponds to the geological features of the region. Thus, the accumulation of moraine on the glaciers of the volcanic peak through erosion of slopes and landslides can be neglected, it is considered to be the bottom moraine, thrown up along the shear planes, the main source of surface moraine on the glaciers of Elbrus. Hence, the debris-cover source in the model is specified to be the result of bedrock erosion rather than slope erosion. The paper discusses calibration processes that allow using simple modeling methods, such as the temperature index method for calculating the surface mass balance, and to simulate the real behavior of glaciers. Despite the fact that the validation of the model revealed a slight underestimation of mass loss at the beginning of the XXI century, the general patterns of mass loss are reproduced correctly, although the energy balance has not been explicitly described. Thus, the adjustment of the model ensures its adaptation to the glaciation conditions on Elbrus.

Probable scenarios of future changes in the Elbrus glaciers and associated with them phenomena such as formation of glacial lakes and remaining ice masses buried under the debris cover are considered. The SSP scenarios (SSP1–1.9, SSP1–2.6, SSP2–4.5, SSP3–7.0, SSP5–8.5) were used for of future climate forcing. Glacier dynamics was simulated using the GloGEMflow model, which was improved by including a module of evolving debris cover. According to the prognostic calculations of the surface mass balance of the glaciers, the loss of ice mass on the Elbrus will accelerate until the end of the 2030s, reaching approximately –1.1±0.3 m w. e. yr.^–1. The volume of the glacier ice is expected to be reducing almost linearly until about 2040, after which the mass loss rate will slow down. Under the warmest climate change scenarios (SSP5–8.5, SSP3–7.0), almost all of the remaining ice masses in the North Caucasus will be concentrated on Elbrus by the end of the century. At the same time, by 2100 the glaciers of Elbrus themselves will retreat up to 4000 m above sea level and higher. In case of moderate warming (SSP1–1.9, SSP1–2.6) the position of glacier fronts may be stabilized at an altitude of 3600–3700 m. The study concerns also the dynamics of the debris cover, predicting its doubling in area and average thickness of 0.22 m by 2040. Although the effect of the debris cover on the total volume of ice on Elbrus is estimated to be minimal, it can temporarily slow down melting of the frontal parts and areas of dead (remaining) ice. According to our estimates, the retreat of the Elbrus glaciers may result in formation of up to 17 new lakes, of which six may potentially be temporarily dammed by dead (remaining) ice zones (up to 60 m thick for Djikaugenkioz). It is expected that the largest lake may be formed on the Djikaugenkioz plateau, it will be dammed by moraine with ice buried under it in the period from 2035 to 2045 if no sufficiently efficient runoff channels will appear. The approximate time and place of formation of such ice masses near the sites of lake formation, depending on the climatic scenario, are shown in the paper, since it is important from the point of view of the risk of outburst floods in the 21st century. Under moderate warming (scenario SSP1–2.6), up to 8 lakes are likely to be formed at the site of retreating glaciers Ulluchiran, Djikaugenkioz, and Bolshoy Azau. All of them may appear in the first half of the century, regardless of the climatic scenario.

Projecting the global evolution of glaciers is crucial to quantify future sea-level rise and changes in glacierfed rivers. Recent intercomparison efforts have shown that a large part of the uncertainties in the projected glacier evolution is driven by the glacier model itself and by the data used for initial conditions and calibration. Here, we quantify the effect that mass balance observations, one of the most crucial data sources used in glacier modelling, have on glacier projections. For this, we model the 21st century global glacier evolution under Coupled Model Intercomparison Phase 6 project (CMIP6) climate scenarios with the Global Glacier Evolution Model (GloGEM) calibrated to match glacierspecific mass balance observations, as opposed to relying on regional mass balance observations. We find that the differences in modelled 21st century glacier changes can be large at the scale of individual glaciers (up to several tens of percent), but tend to average out at regional to global scales (a few percent at most). Our study thus indicates that the added value of relying on glacier-specific observations is at the subregional and local scale, which will increasingly allow projecting the glacier-specific evolution and local impacts for every individual glacier on Earth. To increase the ensemble of models that project global glacier evolution under CMIP6 scenarios, simulations are also performed with the Open Global Glacier Model (OGGM). We project the 2015- 2100 global glacier loss to vary between 25_15% (Glo- GEM) and 29_14%(OGGM) under SSP1-2.6 to 46_26% and 54_29% under SSP5-8.5 (ensemble median, with 95% confidence interval; calibration with glacier-specific observations). Despite some differences at the regional scale and a slightly more pronounced sensitivity to changing climatic conditions, our results agree well with the recent projections by Rounce et al. (2023), thereby projecting, for any emission scenario, a higher 21st century mass loss than the current community estimate from the second phase of the Glacier Model Intercomparison Project (GlacierMIP2)..

Global glacier mass loss has accelerated, producing more and larger glacial lakes. Many of these glacial lakes are a source of glacial lake outburst floods (GLOFs), which pose threats to people and infrastructure. In this Review, we synthesize global changes in glacial lakes and GLOFs. More than 110,000 glacial lakes currently exist, covering a total area of ~15,000 km², having increased in area by ~22% dec^–1 from 1990 to 2020. More than 10 million people are exposed to the impacts of GLOFs, commonly associated with dam failure or wave overtopping associated with mass movements. Although data limitations are substantial, more than 3,000 GLOFs have been recorded from 850 to 2022, particularly in Alaska (24%), High Mountain Asia (HMA; 18%) and Iceland (19%), the majority (64.8%) being from ice-dammed lakes. Recorded GLOFs have increased in most glaciated mountain regions of the world, with ongoing deglaciation and lake expansion expected to increase GLOF frequency further. In HMA, GLOF hazards are projected to triple by 2100, but changes in other regions will likely be lower given topographic constraints on lake evolution. Future research should prioritize acquiring field data on lake and dam properties, producing globally coordinated multi-temporal lake mapping, and robust and efficient modelling of GLOFs for comprehensive hazard assessment and response planning.

Glacier shrinkage and the development of post-glacial ecosystems related to anthropogenic climate change are some of the fastest ongoing ecosystem shifts, with marked ecological and societal cascading consequences. Yet, no complete spatial analysis exists, to our knowledge, to quantify or anticipate this important changeover. Here we show that by 2100, the decline of all glaciers outside the Antarctic and Greenland ice sheets may produce new terrestrial, marine and freshwater ecosystems over an area ranging from the size of Nepal (149,000 ± 55,000 km²) to that of Finland (339,000 ± 99,000 km²). Our analysis shows that the loss of glacier area will range from 22 ± 8% to 51 ± 15%, depending on the climate scenario. In deglaciated areas, the emerging ecosystems will be characterized by extreme to mild ecological conditions, offering refuge for cold-adapted species or favouring primary productivity and generalist species. Exploring the future of glacierized areas highlights the importance of glaciers and emerging post-glacial ecosystems in the face of climate change, biodiversity loss and freshwater scarcity. We find that less than half of glacial areas are located in protected areas. Echoing the recent United Nations resolution declaring 2025 as the International Year of Glaciers’ Preservation and the Global Biodiversity Framework, we emphasize the need to urgently and simultaneously enhance climate-change mitigation and the in situ protection of these ecosystems to secure their existence, functioning and values.

Accelerating glacier melt rates were observed during the last decades. Substantial ice loss occurs particularly during heat waves that are expected to intensify in the future. Because measuring and modelling glacier mass balance on a daily scale remains challenging, short-term mass balance variations, including extreme melt events, are poorly captured. Here, we present a novel approach based on computer-vision techniques for automatically determining daily mass balance variations at the local scale. The approach is based on the automated recognition of colour-taped ablation stakes from camera images and is tested and validated at six stations installed on three Alpine glaciers during the summers of 2019-2022. Our approach produces daily mass balance with an uncertainty of ±0.81 cm w.e. d-1, which is about half of the accuracy obtained from visual readouts. The automatically retrieved daily mass balances at the six sites were compared to average daily mass balances over the last decade derived from seasonal in situ observations to detect and assess extreme melt events. This allows analysing the impact that the summer heat waves which occurred in 2022 had on glacier melt. Our results indicate 23 d with extreme melt, showing a strong correspondence between the heat wave periods and extreme melt events. The combination of below-average winter snowfall and a suite of summer heat waves led to unprecedented glacier mass loss. The Switzerland-wide glacier storage change during the 25 d of heat waves in 2022 is estimated as 1.27 ± 0.10 km3 of water, corresponding to 35 % of the overall glacier mass loss during that summer. The same 25 d of heat waves caused a glacier mass loss that corresponds to 56 % of the average mass loss experienced over the entire melt season during the summers 2010-2020, demonstrating the relevance of heat waves for seasonal melt.

Spatio-temporal reconstruction of winter glacier mass balance is important for assessing long-term impacts of climate change. However, high-altitude regions significantly lack reliable observations, which is limiting the calibration of glaciological and hydrological models. Reanalysis products provide estimates of snow precipitation also for remote high-mountain regions, but this data come with inherent uncertainty, and assessing their biases is difficult given the low quantity and quality of available (long-term) in situ observations. In this study, we aim at improving knowledge on the spatio-temporal variations in winter glacier mass balance by exploring the combination of data from reanalyses and direct snow accumulation observations on glaciers using machine learning. We use the winter mass balance data of 95 glaciers distributed over the European Alps, western Canada, Central Asia and Scandinavia and compare them with the total precipitation from the ERA5 and the MERRA-2 reanalysis products during the snow accumulation seasons from 1981 until 2019. We develop and apply a machine learning model to adjust the precipitation from the reanalysis products along the elevation profile of the glaciers and consequently to reconstruct the winter mass balance in both space (for glaciers without observational data) and time (filling observational data gaps). The employed machine learning model is a gradient boosting regressor (GBR), which combines several meteorological variables from the reanalyses (e.g. air temperature, relative humidity) with topographical parameters. These GBR-derived estimates are evaluated against the winter mass balance data using (i) independent glaciers (site-independent GBR) and (ii) independent accumulation seasons (season-independent GBR). Both approaches resulted in reduced biases and increased correlation between the precipitation of the original reanalyses and the winter mass balance data of the glaciers. Generally, the GBR models have also shown a good representation of the spatial (vertical elevation intervals) and temporal (years) variability of the winter mass balance on individual glaciers.

Snowpack is an important temporal water storage for downstream areas, a potential source of natural hazards (avalanches or floods) and a prerequisite for winter tourism. Here, we use thousands of manual measurements of the water equivalent of the snow cover (SWE) from almost 30 stations between 1,200 and 2,900 m a.s.l. from four long-term monitoring programs (earliest start in 1937) in the center of the European Alps to derive daily SWE based on snow depth data for each station. The inferred long-term daily SWE time series were analyzed regarding spatial differences, as well as potential temporal changes in variability and seasonal averages during the last 7 decades (1957-2022). The investigation based on important hydro-climatological SWE indicators demonstrates significant decreasing trends for mean SWE (Nov-Apr) and for maximum SWE, as well as a significantly earlier occurrence of the maximum SWE and earlier disappearance of the continuous snow cover. The anomalies of mean SWE revealed that the series of low-snow winters since the 1990s is unprecedented since the beginning of measurements. Increased melting during the accumulation period below 2000 m a.s.l is also observed-especially in the most recent years–as well as slower melt rates in spring, and higher day-to-day variability. For these trends no regional differences were found despite the climatological variability of the investigated stations. This indicates that the results are transferable to other regions of the Alps.

Calculation of the calving loss of tidewater glaciers depends on accurate bedrock information. In regional to global-scale projections of future tidewater glacier evolution this dependence is problematic. Bedrock topographies are often unknown and can only be modelled from surface properties. Existing approaches, however, mostly underestimate the ice thickness towards the calving fronts of marine-terminating glaciers. This implies a compromised performance of global-scale projection models which often employ functions of water depth at the calving fronts of tidewater glaciers. Here, we present a sensitivity study that analyses the impact of five different bedrock datasets on projected mass losses from the tidewater glacier Hansbreen in southern Svalbard. Our modelling study calculates the glacier's response to artificial mass-balance forcing. We show that bedrock inaccuracies may lead to a substantially deviating retreat behaviour. The common underestimation of frontal ice thickness/water depth in the modelled bedrock datasets induces an underestimation of sea level-relevant mass losses over the first several decades of modelling. The duration of this period is reduced when assuming warmer climates. Our results thus underline the importance of accurate bedrock topography data for the reliability of glacier evolution projections and for the accuracy of the temporal trajectories of related sea level-relevant mass losses.

More than 13% of the area of the Caucasus glaciers is covered by debris affecting glacier mass balance. Using the Caucasus as example, we introduce a new model configuration that incorporates a physically-based subroutine for the evolution of supraglacial debris into the Global Glacier Evolution Model (GloGEMflow), enabling its application at a regional level. Temporal evolution of debris cover is coupled to glacier dynamics allowing the thickest debris to accumulate in the areas with low velocity. The future evolution of glaciers in the Northern Caucasus is assessed for five Shared Socioeconomic Pathways (SSP) and significance of explicitly incorporating debris-cover formulation in regional glacier modeling is evaluated. Under the more aggressive scenarios, glaciers are projected to disappear almost entirely except on Mount Elbrus, which reaches 5,642 m above sea level, by 2,100. Under the SSP1-1.9 scenario, glacier ice volume stabilizes by 2040. This finding stresses the importance of meeting the Paris Climate Agreement goals and limiting climatic warming to 1.5 °C. We compare evolution of glaciers in the Kuban (more humid western Caucasus) and Terek (drier central and eastern Caucasus) basins. In the Kuban basin, ice loss is projected to proceed at nearly double the rate of that in the Terek basin during the first half of the 21st century. While explicit inclusion of debris cover in modeling leads to a less pronounced projected ice loss, the maximum differences in glacier length, area, and volume occur before 2,100, especially for large valley glaciers diminishing towards the end of the century. These projections show that on average, fraction of debris-covered ice will increase while debris cover will become thinner towards the end of the 21st particularly under the more aggressive scenarios. Overall, the explicit consideration of debris cover has a minor effect on the projected regional glacier mass loss but it improves the representation of changes in glacier geometry locally.

Ice-marginal moraines are widely used as indicators of palaeoclimate in mountainous regions, yet the genetic processes of their formation remain incompletely understood. Here, we present new data on the geomorphology and sedimentology of annually formed moraine ridges in the foreland of Gornergletscher, Switzerland, with the aim of reconstructing the processes of their formation, assessing their preservation potential over longer time scales, and evaluating the climatic significance of these terrestrial archives. A specific focus is set on moraine ridges that formed between 2007 and 2019, a period when the glacier front was subject to pronounced retreat and thinning. Four dominant mechanisms of moraine formation could be identified to be operating across the foreland: (I) freeze-on of submarginal sediments to the advancing glacier sole; (II) the formation of ice-cored moraines controlled by the emergence of englacial debris bands on the glacier surface; (III) efficient bulldozing and deformation of pre-existing fluvial or lacustrine sediments by the advancing glacier; and (IV) inefficient bulldozing of these deposits with the incorporation of dead ice into moraine bodies. The spatial distribution of these processes in the foreland depends on a set of climatological, topographical, and glaciological boundary conditions, the most important of which appears to be the slope of the ice margin. The largest and most well-defined moraines can be genetically linked to efficient bulldozing operating along a sufficiently steep glacier front. These moraines frequently form parts of longer, continuous chains and can be used to calculate frontal retreat rates of the glacier through time. Comparing this record of ice retreat with climatic data reveals a statistically significant correlation to annual air temperature anomalies, which is especially strong for the period between 2005 and 2017. Longer series of annual moraines may therefore provide important information about the climatic drivers that govern glacier retreat. However, dead ice is often found to be incorporated into moraine bodies along the present, thin ice margin. Upon melt-out of the ice, the morphologies and internal structures of these moraines are strongly altered, thus limiting their preservation potential over longer timescales. In such a scenario, the morphological record remaining in a deglaciated landscape may be incomplete or entirely lacking. Therefore, we conclude that great care must be taken when interpreting moraine sequences in a palaeoclimatic context.

Glacier mass loss affects sea level rise, water resources, and natural hazards. We present global glacier projections, excluding the ice sheets, for shared socioeconomic pathways calibrated with data for each glacier. Glaciers are projected to lose 26 ± 6% (+1.5°C) to 41 ± 11% (+4°C) of their mass by 2100, relative to 2015, for global temperature change scenarios. This corresponds to 90 ± 26 to 154 ± 44 millimeters sea level equivalent and will cause 49 ± 9 to 83 ± 7% of glaciers to disappear. Mass loss is linearly related to temperature increase and thus reductions in temperature increase reduce mass loss. Based on climate pledges from the Conference of the Parties (COP26), global mean temperature is projected to increase by +2.7°C, which would lead to a sea level contribution of 115 ± 40 millimeters and cause widespread deglaciation in most mid-latitude regions by 2100.