Glacier retreat is changing the Swiss Alps and creating new ice free areas. Some of these areas could become future reservoirs for renewable electricity and seasonal storage.
This website explores six selected glacier sites and shows how they could contribute to Switzerland's future energy system up to 2100. Explore the map to compare the sites, understand the energy potential and learn about the link to climate change, infrastructure and sustainable development.
01At a Glance
Sites Studied
6
Periglacial reservoir sites across Switzerland
Annual Electricity Production
931 GWh
Combined annual electricity production (G) across all sites
Projection Period
85 yrs
2016–2100 under RCP 2.6 and RCP 8.5 scenarios
Community Equivalent
147k
People supplied with clean electricity per year
02Explore
01
About the Project
Background, context, and the motivation driving this research.
Background
Switzerland's energy transition depends on reliable renewable electricity and long term storage. Hydropower is already the most important domestic renewable energy source in Switzerland, but climate change is changing the Alpine environment. As glaciers retreat, new ice free basins and proglacial lakes may emerge.
Some of these locations could be used as future reservoirs and may support seasonal electricity storage. Ehrbar et al. show that such reservoirs are especially relevant because they can shift energy into the critical winter period, store energy and provide flexible electricity supply.
SDG Alignment
This project focuses on SDG 7, 9, 11 and 13. SDG 7 (Affordable and Clean Energy) is central because the project investigates whether future reservoirs in glacier retreat areas could contribute to a reliable and renewable energy supply. SDG 9 (Industry, Innovation and Infrastructure) is relevant because such reservoirs would require resilient high alpine infrastructure, technical planning and innovation. SDG 11 (Sustainable Cities and Communities) is connected because future energy security also supports resilient settlements and communities that depend on stable electricity supply. SDG 13 (Climate Action) is important because glacier retreat is a direct consequence of climate change and creates difficult adaptation choices for Switzerland.
Motivation
This project is built on the hydropower assessment by Ehrbar et al. (2018), who evaluated potential new hydropower sites in the Swiss periglacial environment against economic, environmental and social criteria. Their study identifies seven best rated sites, but this project focuses on six selected glaciers: Aletsch, Gorner, Grindelwald, Hüfi, Rhône and Roseg. Trift Glacier is not included in the analysis. Based on production values in Ehrbar et al.'s Table 2, these six sites have a combined estimated annual electricity production of 931 GWh, roughly 0.93 TWh per year.
This value should be understood as an estimate of potential, not as a guaranteed future contribution. Ehrbar et al. show that even the best rated sites still have significant drawbacks, and that results depend on uncertainties such as climate development, runoff projections, reservoir operation, technical assumptions and environmental restrictions.
A further challenge is sedimentation. Sediment input can reduce reservoir volume, cause production losses and affect safe operation by blocking outlet structures. Sustainable use therefore depends not only on possible electricity production, but also on whether sedimentation can be understood and managed over the long term.
This map brings these aspects together by showing the six selected glacier retreat areas in a spatial and comparable way. Its goal is to make visible where future reservoir potential could emerge, how large the estimated contribution could be, and why this potential must be considered together with climate change, seasonal storage needs and long term Alpine planning.
Research Question
"To what extent could new reservoirs, located at retreating glaciers, contribute to the expected future energy consumption in Switzerland up to 2100?"
03
Data & Methods
How the data was collected, processed, and visualized.
Data Sources
Surface elevation data comes from the swissALTI3D 2 m digital elevation model (swisstopo), downloaded tile by tile via the geodownloader QGIS plugin for each of the six study sites: Gornergletscher, Aletschgletscher, Grindelwaldgletscher, Hüfigletscher, Rhonegletscher, and Roseggletscher. To model the topography beneath the glacier ice, which is necessary for estimating future lake basin geometry, the Glacier Bed DEM from Farinotti et al. (2019), available via the ETH Research Collection, was used as a complementary elevation source.
Glacier outlines for the 2016 baseline are taken from the Swiss Glacier Inventory 2016 (SGI2016), distributed by GLAMOS, and exported from QGIS as GeoJSON. Future glacier area projections (2000–2100) were obtained from the PyGEM/OGGM model output hosted by NASA NSIDC, provided as NetCDF files containing annual area values for both RCP 2.6 and RCP 8.5 scenarios across ten model ensemble members.
Dam locations, heights, and reservoir specifications for all six sites are based on Ehrbar et al. (2018), a PhD thesis from the ETH Research Collection, supplemented by publicly available information on the Grande Dixence / Gornerli project. Georeferenced figures from the same publication were used to define the alignment of dam walls in the terrain.
Methodology
Dam wall modelling. Dam wall polygons were constructed by georeferencing figures from Ehrbar et al. (2018) and tracing a curved line slightly above the maximum dam height for each site, guided by contour lines from swissALTI3D to ensure the wall connects cleanly with the surrounding terrain. A buffer of 4 m on each side (8 m total width) with flat, capped ends was applied to produce the final wall polygon.
Lake fill simulation. For each site, swissALTI3D tiles overlapping the future lake and dam wall extents were merged into a single raster. The Glacier Bed DEM was then merged into this surface using a minimum-value rule: wherever the glacier bed recorded a lower elevation than the surface, its value replaced the surface, effectively opening the subglacial basin. The dam wall was rasterized and a horizontal distance raster computed. A concave mathematical function was applied to the distance raster to generate a dam wall DEM that is narrow at the crest and widens toward the base, matching the geometry of a realistic arch or gravity dam. This wall DEM was merged into the combined terrain. Using the GRASS tool r.lake, with the lowest basin point as the seed and the dam's maximum supply level as the target, five intermediate fill stages were calculated by dividing the total fill height into equal steps. At each step the flooded raster was polygonized, simplified, and smoothed, a process fully automated via a Python script in the QGIS console. The resulting polygons were subsequently reprojected from the Swiss coordinate system (EPSG:2056) to WGS84 and simplified at a 20 m tolerance for use in the web map.
Glacier retreat animation. Annual glacier area projections were extracted from the PyGEM/OGGM NetCDF files by matching each glacier to its RGI identifier. A boundary erosion algorithm was then applied to the SGI2016 baseline outline: for each year, marginal cells with the lowest elevations are progressively removed until the remaining pixel count matches the PyGEM area fraction for that year, producing a GeoJSON outline per year (2016–2100) for each glacier and scenario.
Representational limitations. It is important to note that the animated glacier outlines are area preserving approximations, not physically or glaciologically accurate simulations of retreat. The boundary erosion approach redistributes ice loss uniformly from the lowest-elevation margins inward, which is a reasonable first order proxy but does not model the actual physical processes governing glacier dynamics, such as ice flow, differential mass balance, englacial temperatures, or the feedback between surface albedo and melt rate. Real glaciers retreat in spatially heterogeneous patterns driven by aspect, slope, shading, and debris cover, none of which are captured here. The outlines should therefore be understood as a spatial representation of projected areal change that communicates the scale and pace of retreat under each climate scenario, rather than as a prediction of where exactly the ice margin will be in any given year.
Data Summary
Elevation ModelswissALTI3D 2m
Glacier ProjectionsPyGEM / OGGM
Glacier OutlinesSGI2016 / GLAMOS
Dam SpecificationsEhrbar et al. 2018
ScenariosRCP 2.6 · RCP 8.5
Time Period2016 – 2100
Glacier Sites6
SoftwareQGIS · Python · Leaflet.js
04
Findings
Hydropower potential of six periglacial reservoir sites in Switzerland, based on Ehrbar et al. (2018).
01The Starting Point: Glaciers in Retreat
Switzerland's glaciers are not a stable backdrop to this analysis; they are disappearing within measurement. The six study glaciers have together shed enormous volumes of ice within the era of systematic observation. Aletschgletscher alone has lost roughly 3.93 km³ of ice since 1980, and Gornergletscher about 2.01 km³ since 1981. These losses were measured directly by comparing repeated digital elevation models of the ice surface (GLAMOS, 2024). It is precisely this retreat that exposes the bedrock basins where future reservoirs could one day sit. The same process that signals a warming climate also creates the spaces this project evaluates.
Aletsch
3.93 km³
Ice volume lost since 1980
Gorner
2.01 km³
Ice volume lost since 1981
Rhone
0.70 km³
Ice volume lost since 1980
02The Policy Driver: Energy Strategy 2050
The interest in periglacial reservoirs is not purely academic; it is anchored in national policy. Following the 2011 Fukushima accident, Switzerland resolved to phase out nuclear power, which had supplied a substantial share of its electricity. The resulting Energy Strategy 2050 commits the country to replacing that capacity largely through renewables and to expanding hydropower storage to manage the transition. Within this frame, Ehrbar et al. (2018) estimate that the best rated periglacial sites could deliver roughly 1.1 TWh of additional annual hydropower production relative to 2016, enough, in principle, to meet the interim hydropower expansion target set for 2035. The six glaciers examined here represent the core of that potential.
03From 1,576 Glaciers to Six Sites
The six glaciers on this map are not an arbitrary selection; they are the survivors of a rigorous filtering process. Ehrbar et al. (2018) began with all 1,576 glaciers in the Swiss Alps. Of these, 62 were identified as technically suitable for a new periglacial hydropower plant, based on having sufficient annual runoff and a usable reservoir basin. These 62 candidates were then scored against an evaluation matrix of 16 criteria spanning three dimensions: economic factors (such as construction cost, installed capacity, and energy yield), environmental factors (such as protected-area conflicts, ecological sensitivity, and landscape impact), and social factors (such as land use, hazards, and acceptance). Only the seven highest-rated sites emerged from this assessment. This project visualizes six of them; Trift was excluded.
The crucial point is that these sites rank highly not merely because they could produce the most electricity, but because they balance energy potential against environmental and social cost. This is what distinguishes a viable site from a merely powerful one.
Starting point
1,576
Glaciers in the Swiss Alps
Technically suitable
62
Sites with sufficient runoff and usable basin
Best-rated sites
7 (6)
Highest scoring across all 16 criteria (6 shown here)
Sites studied
6
Periglacial reservoir sites across Switzerland
Peak power output (W)
310 MW
Combined across all six sites
Annual electricity production (G)
931 GWh
Peak power output × 3,000 full load hours/yr
Storage energy equivalent (E)
1,081 GWh
Energy producible from full reservoir over entire cascade
04Production Specifications
The table below summarises the engineering parameters and energy output estimates for each site as assessed by Ehrbar et al. (2018). Vw is the usable reservoir volume, Qd the design flow, hmax the maximum dam height, zturb the turbine elevation, W the peak power output, G the annual electricity production (installed capacity × 3,000 full load hours/yr), and E the storage energy equivalent (energy producible from the full reservoir volume over the entire downstream cascade).
Glacier
Vwhm³
Qdm³/s
hmaxm
zturbm a.s.l.
WMW
GGWh/yr
EGWh/yr
Aletschgletscher
309
28.6
200
1,445
73
218
396
Gornergletscher
199
18.4
140
1,650
78
235
199
Grindelwaldgletscher
94
8.7
160
950
28
85
64
Hüfigletscher
44
4.1
140
520
35
105
86
Rhonegletscher
56
5.2
100
1,750
19
57
75
Roseggletscher
96
8.9
120
1,000
77
231
261
Total
-
-
-
-
310
931
1,081
05Energy at Community Scale
To contextualize the annual electricity production (G), each site's output is expressed as the number of people it could supply year round, using the Swiss average per capita electricity consumption of 6.31 MWh/person/year (BFE, Schweizerische Elektrizitätsstatistik 2023). A comparable Swiss municipality is shown for reference.
Rhonegletscher
≈ Visp, Valais (8,971)
~9,000 people
57 GWh/yr
Grindelwaldgletscher
≈ Davos (12,700)
~13,500 people
85 GWh/yr
Hüfigletscher
≈ Solothurn (16,802)
~16,600 people
105 GWh/yr
Aletschgletscher
≈ Neuchâtel (33,455)
~34,500 people
218 GWh/yr
Roseggletscher
≈ Sion (37,154)
~36,600 people
231 GWh/yr
Gornergletscher
≈ Schaffhausen (38,982)
~37,200 people
235 GWh/yr
Based on annual electricity production (G) and Swiss average per capita consumption of 6.31 MWh/yr. Source: Ehrbar et al. (2018); BFE, Schweizerische Elektrizitätsstatistik 2023.
06Limitations and Trade-offs
Sedimentation
Periglacial catchments deliver heavy sediment loads, and reservoirs trap that sediment, gradually reducing storage volume and threatening the safe operation of outlet structures. This is not a marginal concern: Ehrbar and colleagues devoted dedicated field campaigns to it, measuring suspended sediment transport in three existing periglacial reservoirs (Lac de Mauvoisin, Griessee, and Gebidem) to better understand how quickly new basins might fill. Sustainable use depends on whether this sedimentation can be managed over decades.
Deep Uncertainty
The production estimates rest on climate projections, runoff models, reservoir operation assumptions, and engineering parameters, each carrying its own uncertainty. The numbers should be read as a well founded order of magnitude, not a forecast.
Downstream Ecology
Building reservoirs in newly ice free terrain is not ecologically neutral. Glacier fed alpine streams host specialized, cold adapted, and often endemic species whose survival is tied to the particular temperature and flow regime that meltwater provides. As glacial influence declines and flows are regulated by dams, these communities are among the most vulnerable to change (Brown et al., 2007). A reservoir that smooths the seasonal flow can erase precisely the conditions these species depend on downstream.
Synthesis
Together, the six periglacial sites could deliver 931 GWh of annual electricity production, enough to supply an estimated ~147,000 people with electricity every year.
The potential is distributed unevenly: Gornergletscher, Roseggletscher, and Aletschgletscher each alone could power a midsized Swiss city, while the smaller sites contribute meaningfully at a municipal scale. Switzerland is not alone in this. The same retreat that is exposing reservoir basins in the Alps is unfolding across the world's glacierized mountains; a global assessment by Farinotti et al. (2019) found that future ice free basins could offer meaningful water storage and hydropower potential well beyond Switzerland, situating the Swiss case within a far larger picture of climate driven landscape change.
These findings speak directly to the goals that motivate the project, above all clean energy (SDG 7) and climate adaptation (SDG 13), but they do so with a deliberate double edge: the very disappearance of the glaciers is what creates the opportunity. That tension is the core message of this map.
05
Sources
References, data sources, and further reading.
Data Sources
GLAMOS – Swiss Glacier Monitoring and Inventory Service
Swiss Glacier Inventory 2016 (SGI2016): glacier outlines used as 2016 baseline; morphometric attributes (area, length, elevation range) used for glacier info panel. Annual glacier monitoring reports used for terminus elevation values. glamos.ch
GLAMOS – Swiss Glacier Volume Change (2025)
Geodetic ice volume change from DEM comparison, used for the ice volume lost figures in the glacier info panel. GLAMOS (2024), release 2025. doi:10.18750/volumechange.2025.r2025. glamos.ch
swisstopo – swissALTI3D
2 m resolution digital elevation model covering Switzerland, used for terrain and dam wall modelling. swisstopo.admin.ch
ETH Research Collection – Glacier Bed DEM
Subglacial topography dataset (Farinotti et al., 2019), used to model future lake basin geometry beneath glacier ice. research-collection.ethz.ch
NASA NSIDC – PyGEM glacier projections
Annual glacier area projections 2000–2100 for RCP 2.6 and RCP 8.5, 50-member ensemble (Rounce et al., 2023). nsidc.org
OpenStreetMap contributors
Basemap tiles used in the interactive map. openstreetmap.org
BFE – Schweizerische Elektrizitätsstatistik 2023
Swiss electricity statistics 2023, used to derive per capita electricity consumption (56.1 TWh ÷ 8.9 M residents ≈ 6.31 MWh/capita/yr). bfe.admin.ch
BFS – Ständige Wohnbevölkerung nach Gemeinde
Swiss permanent resident population by municipality (STAT-TAB), used for the representative town comparison figures. Neuchâtel: Federal Statistical Office. bfs.admin.ch
Academic References
Ehrbar, D., Schmocker, L., Vetsch, D. F. and Boes, R. M. (2018). Hydropower Potential in the Periglacial Environment of Switzerland under Climate Change. Sustainability, 10(8), 2794. https://doi.org/10.3390/su10082794
Ehrbar, D., Schmocker, L., Doering, M., Cortesi, M., Bourban, G., Boes, R. M. and Vetsch, D. F. (2018). Continuous Seasonal and Large Scale Periglacial Reservoir Sedimentation. Sustainability, 10(9), 3265. https://doi.org/10.3390/su10093265
Farinotti, D., Huss, M., Fürst, J. J., Landmann, J., Machguth, H., Maussion, F. and Walters, A. (2019). A consensus estimate for the ice thickness distribution of all glaciers on Earth. Nature Geoscience, 12(3), 168–173. https://doi.org/10.1038/s41561-019-0300-3
Rounce, D. R., Hock, R., Maussion, F., Hugonnet, R., Kochtitzky, W., Huss, M., Berthier, E., Brinkerhoff, D., Compagno, L., Copland, L., Farinotti, D., Menounos, B. and McNabb, R. W. (2023). Global glacier change in the 21st century: Every increase in temperature matters. Science, 379(6627), 78–83. https://doi.org/10.1126/science.abo1324
Brown, L. E., Hannah, D. M. and Milner, A. M. (2007). Vulnerability of alpine stream biodiversity to shrinking glaciers and snowpacks. Global Change Biology, 13(5), 958–966. https://doi.org/10.1111/j.1365-2486.2007.01341.x
United Nations. Sustainable Development Goals 7, 9, 11 and 13. https://sdgs.un.org/goals
06
About
The authors and project behind this website.
Authors
Pascal Andreas Heiniger
University of Zürich
Email
Pascalandreas.heiniger@uzh.ch
Niklas Nowak
University of Zürich
Email
Niklas.nowak@uzh.ch
Yannik Bolli
University of Zürich
Email
Yannik.bolli@uzh.ch
Maximilian Lengenfelder
University of Zürich
Email
Maximilian.lengenfelder@uzh.ch
University
University of Zürich
Module
GEO484 Geovisualization
Semester
Spring Semester 2026
Declarations
Artificial intelligence tools were used to support the website development process, for programming assistance and technical guidance. All content, structure, analysis, and final decisions were reviewed and completed by the authors. All spatial data, glacier projections, and hydropower estimates used in this project were independently gathered, prepared, and processed by the authors. Data sources and processing methods are documented in full on the Data & Methods and Sources pages.
