Glacier Model Research Articles

New projections of 21st century climate and hydrology for Alaska and Hawaiʻi

In the United States, high-resolution, century-long, hydroclimate projection datasets have been developed for water resources planning, focusing on the contiguous United States (CONUS) domain. However, there are few statewide hydroclimate projection datasets available for Alaska and Hawaiʻi. The limited information on hydroclimatic change motivates developing hydrologic scenarios from 1950 to 2099 using climate-hydrology impact modeling chains consisting of multiple statistically downscaled climate projections as input to hydrologic model simulations for both states. We adopt an approach similar to the previous CONUS hydrologic assessments where: 1) we select the outputs from ten global climate models (GCM) from the Coupled Model Intercomparison Project Phase 5 with Representative Concentration Pathways 4.5 and 8.5; 2) we perform statistical downscaling to generate climate input data for hydrologic models (12-km grid-spacing for Alaska and 1-km for Hawaiʻi); and 3) we perform process-based hydrologic model simulations. For Alaska, we have advanced the hydrologic model configuration from CONUS by using the full water-energy balance computation, frozen soils and a simple glacier model. The simulations show that robust warming and increases in precipitation produce runoff increases for most of Alaska, with runoff reductions in the currently glacierized areas in Southeast Alaska. For Hawaiʻi, we produce the projections at high resolution (1 km) which highlight high spatial variability of climate variables across the state, and a large spread of runoff across the GCMs is driven by a large precipitation spread across the GCMs. Our new ensemble datasets assist with state-wide climate adaptation and other water planning.

GlacierNet2: A hybrid Multi-Model learning architecture for alpine glacier mapping

In recent decades, climate change has significantly affected glacier dynamics, resulting in mass loss and an increased risk of glacier-related hazards including supraglacial and proglacial lake development, as well as catastrophic outburst flooding. Rapidly changing conditions dictate the need for continuous and detailed observations and analysis of climate-glacier dynamics. Thematic and quantitative information regarding glacier geometry is fundamental for understanding climate forcing and the sensitivity of glaciers to climate change, however, accurately mapping debris-cover glaciers (DCGs) is notoriously difficult based upon the use of spectral information and conventional machine-learning techniques. The objective of this research is to improve upon an earlier proposed deep-learning-based approach, GlacierNet, which was developed to exploit a convolutional neural-network segmentation model to accurately outline regional DCG ablation zones. Specifically, we developed an enhanced GlacierNet2 architecture thatincorporates multiple models, automatic post-processing, and basin-level hydrological flow techniques to improve the mapping of DCGs such that it includes both the ablation and accumulation zones. Experimental evaluations demonstrate that GlacierNet2 improves the estimation of the ablation zone and allows a high level of intersection over union (IOU: 0.8839) score. The proposed architecture provides complete glacier (both accumulation and ablation zone) outlines at regional scales, with an overall IOU score of 0.8619. This is a crucial first step in automating complete glacier mapping that can be used for accurate glacier modeling or mass-balance analysis.

Increased mass loss of glaciers in Volcán Domuyo (Argentinian Andes) between 1962 and 2020, revealed by aerial photos and satellite stereo imagery

AbstractWe present geodetic mass-balance estimates for ten glaciers (22.6 ± 1.1 km2) around Volcán Domuyo between 1962 and 2020 (and 46 glaciers covering 29 ± 1.5 km2 between 1984 and 2020), derived from airborne, ASTER and Pléiades imagery. Overall, we find a slightly negative mass balance (−0.15 ± 0.09 m w.e. a–1) for the entire 1962–2020 time span. A closer inspection of sub-periods reveals, however, an increasingly negative mass balance with time. The Domuyo glaciers shifted from a moderately positive mass balance of +0.28 ± 0.13 m w.e. a–1 between 1962 and 1984, to a strongly negative mass balance of −0.99 ± 0.19 m w.e. a–1 between 2012 and 2020. An increase in summer temperatures and a decrease in winter precipitation during the last four decades are likely drivers of the observed glacier changes. We support this finding by implementing a minimal glacier model, relying solely on monthly precipitation and air temperatures. The mass-balance evolution detected in Volcán Domuyo is consistent with other sites spread across the Central Andes, suggesting rapidly increasing glacier wastage impacts are occurring at a geographically wider scale.

Importance of the seasonal temperature and precipitation variation on glacial evolutions during the LGM at monsoonal Himalaya based on Cogarbu valley

Glacier models have played a more and more important role in reconstructing the paleoglacier and paleoclimate, but the detailed analysis based on high spatial and temporal resolution simulations is rare. In this study, we deeply analyzed the paleoglacial evolution on seasonal scale and its relationship with the large-scale circulation shifts during the Last Glacial Maximum (LGM) in Cogarbu valley, a typical monsoon influence region, using the Open Global Glacier Model driven by 12 Global Circulation Models and their average. The paleoglacier reconstructed under the mean climate is closer to the observations, with the glacier length, area, and volume of 19.8 km, 35.7 km2, and 6.6 km3, respectively. The reconstructed equilibrium-line altitude is ~5253 m, about 662 m lower than modern. The temperature has a 4.6 °C decreasing with a 16.3% reduction of precipitation. The seasonal temperature decreases about −4.0, −4.6, −4.2, and − 5.5 °C in spring, summer, autumn, and winter, respectively. Detailed analysis on atmospheric circulation shifts indicates that the high correlation of temperature change between TP and North Atlantic Ocean might be caused by their similar response to the global climate change results from the reduction of insolation rather than baroclinic waves. In summer, southward-shifted and narrowed the Intertropical Convergence Zone weakened the Indian Summer Monsoon, decreasing the summer precipitation of −22.4% compared to modern. In winter, the enhancement of the South Branch of Westerlies increased the winter precipitation of 10.0%. Despite more winter precipitation refueled the glacial accumulation, the decreasing of the summer temperature has dominant influence on the annual net surface mass balance via reducing the summer ablation during the LGM.

The Role of the Snow Ratio in Mass Balance Change under a Warming Climate for the Dongkemadi Glacier, Tibetan Plateau

Abstract Understanding the effects of the snow ratio on glacier mass balance under variable climatic conditions is crucial for predicting how glaciers will respond to climate change, and for forecasting water supplies to surrounding lowland areas. Due to recent climate change, the historical annual snow ratio of the Dongkemadi (DKMD) Glacier showed a significant increasing trend (0.0538% a−1, p < 0.05), and an abrupt upward change in 1977 due to decreasing precipitation concentration. Snow ratios with fixed precipitation concentration and nonwarming climate scenarios were calculated to isolate the impact of the snow ratio on glacier mass balance. Under nonwarming conditions, the snow ratio showed little variability, ranging from 88.4% to 99.9%. Glacier modeling results comparing five snow ratio scenarios from 1961 to 2009 showed three main features as follows. (i) Glacier mass balance was low and more sensitive to a warming climate for lower snow ratio scenarios. (ii) The difference in mass balance between the scenarios fluctuated, but generally increased with time. Spatially, the ablation area change was larger (0.4 km2), and the equilibrium line altitude was higher (5.9 m) in scenarios with lower snow ratios. (iii) The change in net shortwave radiation was the main reason for changes in glacial melt, and the albedo played a key role in controlling the difference of glacier energy balance between snow ratio scenarios. Rain increment only accounted for about 20%–33% of meltwater increment. Overall, this study provides valuable information to evaluate how snow ratios impact the mass balance of glaciers with ongoing climate change.

Glacial runoff buffers droughts through the 21st century

Abstract. Global climate model projections suggest that 21st century climate change will bring significant drying in the terrestrial midlatitudes. Recent glacier modeling suggests that runoff from glaciers will continue to provide substantial freshwater in many drainage basins, though the supply will generally diminish throughout the century. In the absence of dynamic glacier ice within global climate models (GCMs), a comprehensive picture of future hydrological drought conditions in glaciated regions has been elusive. Here, we leverage the results of existing GCM simulations and a global glacier model to evaluate glacial buffering of droughts in the Standardized Precipitation-Evapotranspiration Index (SPEI). We compute one baseline version of the SPEI and one version modified to account for glacial runoff changing over time, and we compare the two for each of 56 large-scale glaciated basins worldwide. We find that accounting for glacial runoff tends to increase the multi-model ensemble mean SPEI and reduce drought frequency and severity, even in basins with relatively little (<2 %) glacier cover. When glacial runoff is included in the SPEI, the number of droughts is reduced in 40 of 56 basins and the average drought severity is reduced in 53 of 56 basins, with similar counts in each time period we study. Glacial drought buffering persists even as glacial runoff is projected to decline through the 21st century. Results are similar under representative concentration pathway (RCP) 4.5 and 8.5 emissions scenarios, though results for the higher-emissions RCP 8.5 scenario show wider multi-model spread (uncertainty) and greater incidence of decreasing buffering later in the century. A k-means clustering analysis shows that glacial drought buffering is strongest in moderately glaciated, arid basins.

Projection of Streamflow Changes Under CMIP6 Scenarios in the Urumqi River Head Watershed, Tianshan Mountain, China

Analyzing climate change impacts on hydrology and future water supply projections is essential for effective water resource management and planning in the large river basins of Asia. In these regions, streamflow and glacier melt remain subject to significant uncertainties due to the lack of confidence in climate change projections and modeling methods. In this study, a glacier dynamics model (the Open Global Glacier Model was coupled with a glacio-hydrological model [the Glacio-hydrological Degree-day Model (GDM)] to predict possible hydrological changes in the head watershed of the Urumqi River under three shared socioeconomic pathways SSP2-4.5, SSP3-7.0, and SSP5-8.5. The GDM was calibrated and validated against in situ observed discharge data for the 2007–2011 and 2012–2018 periods. The resulting Nash–Sutcliffe efficiency (NSE) values were 0.82 and 0.81, respectively. The GDM was driven with an ensemble of five downscaled CMIP6 datasets to examine the potential impacts of climate change on hydrologic processes in the basin. Four runoff components were simulated with the GDM: base flow, rainfall, ice melt, and snow melt. It was determined that rainfall constituted the predominant source of runoff, followed by baseflow and ice melt. During the calibration and validation periods, snow and ice melt contributed 25.14 and 25.62%, respectively, to the total runoff. Under all SSP scenarios, the projected runoff decline indicated that the peak runoff time had passed. It was revealed that a 2°C increase in the monthly average temperature could result in a 37.7% increase in the total discharge of the basin. Moreover, the GDM was more responsive to changes in air temperature than to changes in glacier extent.

Advances in modelling large river basins in cold regions with Modélisation Environmentale Communautaire—Surface and Hydrology (MESH), the Canadian hydrological land surface scheme

AbstractCold regions provide water resources for half the global population yet face rapid change. Their hydrology is dominated by snow, ice and frozen soils, and climate warming is having profound effects. Hydrological models have a key role in predicting changing water resources but are challenged in cold regions. Ground‐based data to quantify meteorological forcing and constrain model parameterization are limited, while hydrological processes are complex, often controlled by phase change energetics. River flows are impacted by poorly quantified human activities. This paper discusses the scientific and technical challenges of the large‐scale modelling of cold region systems and reports recent modelling developments, focussing on MESH, the Canadian community hydrological land surface scheme. New cold region process representations include improved blowing snow transport and sublimation, lateral land‐surface flow, prairie pothole pond storage dynamics, frozen ground infiltration and thermodynamics, and improved glacier modelling. New algorithms to represent water management include multistage reservoir operation. Parameterization has been supported by field observations and remotely sensed data; new methods for parameter identification have been used to evaluate model uncertainty and support regionalization. Additionally, MESH has been linked to broader decision‐support frameworks, including river ice simulation and hydrological forecasting. The paper also reports various applications to the Saskatchewan and Mackenzie River basins in western Canada (0.4 and 1.8 million km2). These basins arise in glaciated mountain headwaters, are partly underlain by permafrost, and include remote and incompletely understood forested, wetland, agricultural and tundra ecoregions. These illustrate the current capabilities and limitations of cold region modelling, and the extraordinary challenges to prediction, including the need to overcoming biases in forcing data sets, which can have disproportionate effects on the simulated hydrology.

Projected 21st-Century Glacial Lake Evolution in High Mountain Asia

In High Mountain Asia (HMA), rising temperatures and retreating glaciers are leading to the formation of new glacial lakes and the expansion of existing ones. The sudden release of water from such lakes can lead to devastating glacial lake outburst floods (GLOF) threatening people and infrastructure for many kilometers downstream. Therefore, information on future glacial lakes, e.g., their location, area and volume as well as the timing of their development, is vital for sustainable development of settlements and infrastructures. In this study, we present comprehensive estimates for future glacial lake development in HMA with unprecedented temporal resolution. We rely on an ensemble of fifteen global climate models using the newest CMIP6 data and employ a set of four Shared Socioeconomic Pathway (SSP) scenarios. With the Open Global Glacier Model (OGGM), we use a modeling framework that explicitly simulates glacier dynamics in order to model glacier change until 2100 and estimate the formation period for each of the 2,700 largest future glacial lakes (>0.1 km2) in HMA. We estimate the glacial lake area in the entire region to grow by 474 ± 121 km2 for SSP126 and 833 ± 148 km2 for SSP585. Following recent estimates of currently existing glacial lakes (>0.1 km2), this would constitute an increase in lake area of ∼120–∼210% in 2100 compared to 2018. The lake volume is expected to increase by 22.8 ± 6.7 km3 for SSP126 and 39.7 ± 7.7 km3 for SSP585. This range includes a drastic tenfold increase in lake volume, from estimated 3.9 km3 in 2018 to 43.6 ± 7.7 km3 in 2100. However, there is a considerable spread between total and relative increase in glacial lake area and volume for different sub-regions of High Mountain Asia. As both, lake area and lake volume, could to lead to an increase in GLOF risk, the results emphasize the urgent need for more localized, in-depth studies at especially vulnerable locations in order to enable local communities to adapt to emerging challenges, to implement risk minimization measures, and to improve sustainable development in High Mountain Asia.

Energy mass balance and flow modeling of early Holocene glaciers in the Queshque valley, Cordillera Blanca, Peru

Our limited knowledge of the timing and pattern of early Holocene climate variability in the tropical Andes hinders our ability to evaluate any potential linkages between low and high latitude oceanic-atmospheric dynamics. There is mounting evidence that glaciers in the Peruvian Andes stabilized at times during the early Holocene, suggesting there were periods of colder and/or wetter conditions that interrupted an overall pattern of a warmer and drier climate. Evaluating the global significance of these glacial fluctuations requires that the nature of these apparent cooling events and any shifts in the hydrologic cycle be better quantified. Here we apply a physically-based glacier model to reconstruct and interpret early Holocene-aged paleoglaciers in steady-state with a range of tropical climatic conditions in the Queshque valley of the Cordillera Blanca, Peru (9.8°S). This model uses a LiDAR-based digital elevation model (DEM) and hourly meteorological data as inputs to calculate the fully-distributed surface energy and mass balance (SEMB). This approach allows us to better capture the diurnal range of environmental variability at a spatial resolution that has not been achieved with previous paleo-glacier modeling efforts in this region. A 3-D rendition of paleoglaciers was then developed based on a flow model that responds to the SEMB input, accounting for the valley topography. The model was validated by simulating glaciers that match both field and satellite observations of modern glacier area limits, as well as ice thicknesses that are consistent with recently measured ground-penetrating radar (GPR) profiles of the main glacier tongue. After adjusting for changes in early Holocene global radiation values, and using regional paleoclimate records as constraints for the model, the maximum early Holocene ice limit in the valley dating to ∼10.8 (±0.1) ka was reconstructed using a 3.0 °C cooling and a 25% increase in precipitation relative to today. An up-valley ice extent dating to ∼9.4 (±0.3) ka was reconstructed using a 2.8 °C cooling and no change in precipitation amounts relative to today. These results suggest that conditions were cold and wet enough to maintain glaciers at times during the early Holocene, even though a shift to drier conditions combined with modestly warmer temperatures drove a phase of ice retreat from at least ∼10.8 to ∼9.4 ka. These results suggest that southern tropical temperatures and the hydrologic cycle rapidly reorganized during the early Holocene in conjunction with diminishing ice sheets and shifting environmental conditions in the high latitudes.

Albedo Parametrizations for the Laohugou Glacier No.12 in the Qilian Mountains—Previous Models and an Alternative Approach

Accurate estimates of albedo can be crucial for energy balance models of glaciers. A number of algorithms exist which are often site dependent and rely on accurate measurements or estimates of snow depth. Using the well-established COSIMA model we simulate the energy and mass balance of the Laohugou Glacier No.12 in the Qilian Mountains, on the northern fringe of the Qinghai-Tibetan Plateau, a glacier that has been well studied in the past. Using energy flux and mass balance measurements between 2010 and 2015 we were able to validate the model over multiple seasons. Using the original albedo parametrization, the model fails to reproduce the observed mass balance. We show that this is due to the failure to estimate snow depth accurately. We therefore applied two alternative albedo algorithms, one well established example and one new parametrization only dependent on temperature and time since last snow fall. As a result, mass balance simulations improve considerably from a RMSE of 0.53 m w.e. for the original parametrization to 0.39 and 0.19 m w.e. for the uncalibrated established and the new calibrated model respectively. Modelled albedo during the ablation period (NSE = 0.05, R2 = 0.33) is more accurate than during the accumulation period (NSE = −0.37, R2 = 0.04). Testing the new model at another glacier on the Tibetan Plateau shows that a local recalibration of parameters remains necessary to achieve satisfying results. Investigations into the effect of impurities in snow, regional moisture sources and changing surface characteristics with rising temperatures will be crucial for accurate projections into the future.

Late Pleistocene glacial chronologies and paleoclimate in the northern Rocky Mountains

Abstract. The geologic record of mountain glaciations is a robust indicator of terrestrial paleoclimate change. During the last glaciation, mountain ranges across the western US hosted glaciers while the Cordilleran and Laurentide ice sheets flowed to the west and east of the continental divide, respectively. Records detailing the chronologies and paleoclimate significance of these ice advances have been developed for many sites across North America. However, relatively few glacial records have been developed for mountain glaciers in the northern Rocky Mountains near former ice sheet margins. Here, we report cosmogenic beryllium-10 surface exposure ages and numerical glacier modeling results, which show that mountain glaciers in the northern Rockies abandoned terminal moraines after the end of the global Last Glacial Maximum around 17–18 ka and could have been sustained by −10 to −8.5 ∘C temperature depressions relative to modern assuming similar or less than modern precipitation. Additionally, we present a deglacial chronology from the northern Rocky Mountains that indicates while there is considerable variability in initial moraine abandonment ages across the Rocky Mountains, the pace of subsequent ice retreat through the late glacial exhibits some regional coherence. Our results provide insight on potential regional mechanisms driving the initiation of and sustained deglaciation in the western US, including rising atmospheric CO2 and ice sheet collapse.

Towards ice-thickness inversion: an evaluation of global digital elevation models (DEMs) in the glacierized Tibetan Plateau

Abstract. Accurate estimates of regional ice thickness, which are generally produced by ice-thickness inversion models, are crucial for assessments of available freshwater resources and sea level rise. A digital elevation model (DEM) derived from surface topography of glaciers is a primary data source for such models. However, the scarce in situ measurements of glacier surface elevation limit the evaluation of DEM uncertainty. Hence the influence of DEM uncertainty on ice-thickness modeling remains unclear over the glacierized area of the Tibetan Plateau (TP). Here, we examine the performance of six widely used and mainly global-scale DEMs: AW3D30 (ALOS – Advanced Land Observing Satellite – World 3D – 30 m; 30 m), SRTM-GL1 (Shuttle Radar Topography Mission Global 1 arc second; 30 m), NASADEM (NASA Digital Elevation Model; 30 m), TanDEM-X (TerraSAR-X add-on for Digital Elevation Measurement, synthetic-aperture radar; 90 m), SRTM v4.1 (Shuttle Radar Topography Mission; 90 m), and MERIT (Multi-Error-Removed Improved-Terrain; 90 m) over the glacierized TP by comparison with ICESat-2 (Ice, Cloud and land Elevation Satellite) laser altimetry data while considering the effects of glacier dynamics, terrain factors, and DEM misregistration. The results reveal NASADEM to be the best performer in vertical accuracy, with a small mean error (ME) of 0.9 m and a root mean squared error (RMSE) of 12.6 m, followed by AW3D30 (2.6 m ME and 11.3 m RMSE). TanDEM-X also performs well (0.1 m ME and 15.1 m RMSE) but suffers from serious errors and outliers on steep slopes. SRTM-based DEMs (SRTM-GL1, SRTM v4.1, and MERIT) (13.5–17.0 m RMSE) had an inferior performance to NASADEM. Errors in the six DEMs increase from the south-facing to the north-facing aspect and become larger with increasing slope. Misregistration of the six DEMs relative to the ICESat-2 footprint in most glacier areas is small (less than one grid spacing). In a next step, the influence of six DEMs on four ice-thickness inversion models – GlabTop2 (Glacier bed Topography), Open Global Glacier Model (OGGM), Huss–Farinotti (HF), and Ice Thickness Inversion Based on Velocity (ITIBOV) – is intercompared. The results show that GlabTop2 is sensitive to the accuracy of both elevation and slope, while OGGM and HF are less sensitive to DEM quality and resolution, and ITIBOV is the most sensitive to slope accuracy. NASADEM is the best choice for ice-thickness estimates over the whole TP.

GlaciDat – a GIS database of submarine glacial landforms and sediments in the Arctic

A digital database for submarine glacial landforms and sediments formed in the Arctic during and since the Last Glacial Maximum was created in order to facilitate and underpin new research on palaeo‐ice sheets and tidewater glacier dynamics. The glacimarine database (GlaciDat) documents and standardises evidence of previous glacial activity as visible on the contemporary seafloor of fjords and continental shelves around Svalbard, Greenland, Alaska, northern Russia and north of 66°30′N in Canada and Norway. An extensive literature search was conducted to create GlaciDat, which compiles nearly 60 000 individual submarine landforms, more than 1000 sediment cores and 232 radiocarbon dates. Glacial landforms included are cross‐shelf troughs, trough‐mouth fans, grounding‐zone wedges, lateral moraines, overridden moraines, (mega‐scale) glacial lineations, drumlins, crag‐and‐tails, medial moraines, terminal moraines, debris‐flow lobes (including glacier‐contact fans), recessional moraines, De Geer moraines, crevasse‐fill ridges, eskers, hill‐hole pairs, crescentic scours, and submarine channels. They were digitised as point, line and polygon features alongside a list of their individual characteristics. Sediment core locations are attributed with a description of the sampled lithofacies and sedimentation rates where available. Landforms and sediments have been standardised according to predefined nomenclatures to make the glacial evidence as consistent as possible. Marine radiocarbon dates were included when thought to be relevant for constraining the timing of large‐scale palaeo‐ice dynamics. Outlines of bathymetric data sets, which have previously been used for glacial geomorphological mapping, were also included to give an overview of already investigated research areas. GlaciDat is available for download (https://doi.pangaea.de/10.1594/PANGAEA.937782) and will aid researchers in the reconstruction of past ice dynamics and the interpretation of Arctic glacial landform–sediment assemblages. Moreover, as well as providing a comprehensive bibliography on Arctic glacial geomorphological and sedimentological research, it is intended to serve as a basis for future modelling of Arctic glacier and ice‐sheet dynamics.

Mid-Term Monitoring of Glacier’s Variations with UAVs: The Example of the Belvedere Glacier

Recently, Unmanned Aerial Vehicles (UAV) have opened up unparalleled opportunities for alpine glacier monitoring, as they allow for reconstructing extensive and high-resolution 3D models. In order to evaluate annual ice flow velocities and volume variations, six yearly measurements were carried out between 2015 and 2020 on the debris-covered Belvedere Glacier (Anzasca Valley, Italian Alps) with low-cost fixed-wing UAVs and quadcopters. Every year, ground control points and check points were measured with GNSS. Images acquired from UAV were processed with Structure-from-Motion and Multi-View Stereo algorithms to build photogrammetric models, orthophotos and digital surface models, with decimetric accuracy. Annual glacier velocities were derived by combining manually-tracked features on orthophotos with GNSS measurements. Velocities ranging between 17 m y−1 and 22 m y−1 were found in the central part of the glacier, whereas values between 2 m y−1 and 7 m y−1 were found in the accumulation area and at the glacier terminus. Between 2 × 106 m3 and 3.5 × 106 m3 of ice volume were lost every year. A pair of intra-year measurements (October 2017–July 2018) highlighted that winter and spring volume reduction was ∼1/4 of the average annual ice loss. The Belvedere monitoring activity proved that decimetric-accurate glacier models can be derived with low-cost UAVs and photogrammetry, limiting in-situ operations. Moreover, UAVs require minimal data acquisition costs and allow for great surveying flexibility, compared to traditional techniques. Information about annual flow velocities and ice volume variations of the Belvedere Glacier may have great value for further understanding glacier dynamics, compute mass balances, or it might be used as input for glacier flow modelling.

Deep learning speeds up ice flow modelling by several orders of magnitude

AbstractThis paper introduces the Instructed Glacier Model (IGM) – a model that simulates ice dynamics, mass balance and its coupling to predict the evolution of glaciers, icefields or ice sheets. The novelty of IGM is that it models the ice flow by a Convolutional Neural Network, which is trained from data generated with hybrid SIA + SSA or Stokes ice flow models. By doing so, the most computationally demanding model component is substituted by a cheap emulator. Once trained with representative data, we demonstrate that IGM permits to model mountain glaciers up to 1000 × faster than Stokes ones on Central Processing Units (CPU) with fidelity levels above 90% in terms of ice flow solutions leading to nearly identical transient thickness evolution. Switching to the GPU often permits additional significant speed-ups, especially when emulating Stokes dynamics or/and modelling at high spatial resolution. IGM is an open-source Python code which deals with two-dimensional (2-D) gridded input and output data. Together with a companion library of trained ice flow emulators, IGM permits user-friendly, highly efficient and mechanically state-of-the-art glacier and icefields simulations.

The Challenge of Non-Stationary Feedbacks in Modeling the Response of Debris-Covered Glaciers to Climate Forcing

Ongoing changes in mountain glaciers affect local water resources, hazard potential and global sea level. An increasing proportion of remaining mountain glaciers are affected by the presence of a surface cover of rock debris, and the response of these debris-covered glaciers to climate forcing is different to that of glaciers without a debris cover. Here we take a back-to-basics look at the fundamental terms that control the processes of debris evolution at the glacier surface, to illustrate how the trajectory of debris cover development is partially decoupled from prevailing climate conditions, and that the development of a debris cover over time should prevent the glacier from achieving steady state. We discuss the approaches and limitations of how this has been treated in existing modeling efforts and propose that “surrogate world” numerical representations of debris-covered glaciers would facilitate the development of well-validated parameterizations of surface debris cover that can be used in regional and global glacier models. Finally, we highlight some key research targets that would need to be addressed in order to enable a full representation of debris-covered glacier system response to climate forcing.

Northeastern Patagonian Glacier Advances (43°S) Reflect Northward Migration of the Southern Westerlies Towards the End of the Last Glaciation

The last glacial termination was a key event during Earth’s Quaternary history that was associated with rapid, high-magnitude environmental and climatic change. Identifying its trigger mechanisms is critical for understanding Earth’s modern climate system over millennial timescales. It has been proposed that latitudinal shifts of the Southern Hemisphere Westerly Wind belt and the coupled Subtropical Front are important components of the changes leading to global deglaciation, making them essential to investigate and reconstruct empirically. The Patagonian Andes are part of the only continental landmass that fully intersects the Southern Westerly Winds, and thus present an opportunity to study their former latitudinal migrations through time and to constrain southern mid-latitude palaeo-climates. Here we use a combination of geomorphological mapping, terrestrial cosmogenic nuclide exposure dating and glacial numerical modelling to reconstruct the late-Last Glacial Maximum (LGM) behaviour and surface mass balance of two mountain glaciers of northeastern Patagonia (43°S, 71°W), the El Loro and Río Comisario palaeo-glaciers. In both valleys, we find geomorphological evidence of glacier advances that occurred after the retreat of the main ice-sheet outlet glacier from its LGM margins. We date the outermost moraine in the El Loro valley to 18.0 ± 1.15 ka. Moreover, a series of moraine-matching simulations were run for both glaciers using a spatially-distributed ice-flow model coupled with a positive degree-day surface mass balance parameterisation. Following a correction for cumulative local surface uplift resulting from glacial isostatic adjustment since ∼18 ka, which we estimate to be ∼130 m, the glacier model suggests that regional mean annual temperatures were between 1.9 and 2.8°C lower than present at around 18.0 ± 1.15 ka, while precipitation was between ∼50 and ∼380% higher than today. Our findings support the proposed equatorward migration of the precipitation-bearing Southern Westerly Wind belt towards the end of the LGM, between ∼19.5 and ∼18 ka, which caused more humid conditions towards the eastern margins of the northern Patagonian Ice Sheet a few centuries ahead of widespread deglaciation across the cordillera.

A fully-coupled 3D model of a large Greenlandic outlet glacier with evolving subglacial hydrology, frontal plume melting and calving

AbstractWe present the first fully coupled 3D full-Stokes model of a tidewater glacier, incorporating ice flow, subglacial hydrology, plume-induced frontal melting and calving. We apply the model to Store Glacier (Sermeq Kujalleq) in west Greenland to simulate a year of high melt (2012) and one of low melt (2017). In terms of modelled hydrology, we find perennial channels extending 5 km inland from the terminus and up to 41 and 29 km inland in summer 2012 and 2017, respectively. We also report a hydrodynamic feedback that suppresses channel growth under thicker ice inland and allows water to be stored in the distributed system. At the terminus, we find hydrodynamic feedbacks exert a major control on calving through their impact on velocity. We show that 2012 marked a year in which Store Glacier developed a fully channelised drainage system, unlike 2017, where it remained only partially developed. This contrast in modelled behaviour indicates that tidewater glaciers can experience a strong hydrological, as well as oceanic, control, which is consistent with observations showing glaciers switching between types of behaviour. The fully coupled nature of the model allows us to demonstrate the likely lack of any hydrological or ice-dynamic memory at Store Glacier.

Glacial Change and Its Hydrological Response in Three Inland River Basins in the Qilian Mountains, Western China

Glacial changes have great effects on regional water security because they are an important component of glacierized basin runoff. However, these impacts have not yet been integrated and evaluated in the arid/semiarid inland river basins of western China. Based on the degree-day glacier model, glacier changes and their hydrologic effects were studied in 12 subbasins in the Shiyang River basin (SYRB), Heihe River basin (HHRB) and Shule River basin (SLRB). The results showed that the glacier area of each subbasin decreased by 16.7–61.7% from 1965 to 2020. By the end of this century, the glacier areas in the three basins will be reduced by 64.4%, 72.0% and 83.4% under the three climate scenarios, and subbasin glaciers will disappear completely after the 2070s even under RCP2.6. Glacial runoff in all subbasins showed a decreasing–increasing–decreasing trend, with peak runoff experienced in 11 subbasins during 1965~2020. The contribution of glacial meltwater to total runoff in the basin ranged from 1.3% to 46.8% in the past, and it will decrease in the future due to increasing precipitation and decreasing glacial meltwater. However, the scale differences in glacier runoff are significant when aggregated over the region/basin/subbasin. This suggests that the results of large-scale generalization may be misleading for subbasin glacier water resource evaluations. Therefore, the hydrological effects of glaciers should be studied more in subbasins to provide an accurate reference for practical water resource management.