FUGE: Glacier retreat – still sufficient water for hydroelectric power production?


Using improved methods, the project analysed 50 glaciers in Switzerland and produced models of how they are melting. The run-offs predicted until 2100 are relevant for power plants. Adaptation strategies for hydropower plants were developed in collaboration with the companies running them.

Project description (Completed research project)

Glaciers and climate are intimately related to each other. Climate changes are known to have a strong impact on the runoff regime of alpine streams. The goals of the project were (1) to provide an assessment of glacier changes in Switzerland and (2) to forecast the evolution of glaciers until 2100, and its impacts on hydropower production and the management of water infrastructure.


First we developed a glacier mass balance and runoff model with improved representation of glacier processes. New methods to model area-wide glacier mass balance were established. We developed methods for the redistribution of winter snow, which is one of the weak components in existing mass balance models. Based on repeated LIDAR data sets derived for Haut Glacier d’Arolla over two different winter seasons, we performed a fractal analysis of the scaling properties of snow depth fields. Glacier melt models of different complexities were compared to determine the best approach for long-term simulations. On this basis, the future evolution of the glaciers in the Swiss Alps was modelled with a coupled mass balance / ice dynamic model. The runoff was modelled with a coupled mass balance / runoff model. In order to improve the accuracy of the initial conditions of spatial ice thickness distribution and bed topography, we used a helicopter based ground penetrating radar (GPR) system. This radar allowed us to measure ice thickness profiles on several glaciers.

The glacier model outputs were used for the economic analysis of changes related to climate and glacier changes. For this, a management model was developed. This model was based on the hypothesis of revenue maximization. It provides future revenue relating to expected runoff and electricity price scenarios. On this basis, an investment analysis was performed with Net Present Value and Real Options Analysis. In addition to the strictly socio-economic analysis, a political science experiment was carried out. Through that analysis we could assess the impact of scientific narratives on political causal stories and on the reaction of stakeholders to new scientific results.


The results of the fractal analysis of our LIDAR records on Haut Glacier d’Arolla show that the spatial pattern of snow accumulation seems constant from one year to the other, even if the snow depth differed in those two years. In our research it became clear that mass balance models of intermediate sophistication perform the best over the long-term and are the most accurate for long-term simulation. Energy-balance models and conceptual temperature-index models are oversensitive to temperature fluctuations and therefore less accurate. The former can work well only when good quality input data is available. For the latter, extensive calibration is necessary. The coupling of such mass balance model with a high order glacier flow model seems to be the most appropriate for long-term modelling of future glacier changes. In the case study in the Mauvoisin catchment we demonstrated the importance of advanced knowledge of glacier geometry and spatial ice volume distribution for accurate glacier evolution and runoff projections.

The results of the economic study show that the impact of climate change can be mitigated through a fine reservoir management. Moreover, we show that uncertainties are mainly related to electricity prices since those are more variable than runoff. We further point out that the present market situation does not justify an investment in a pumped-storage installation in Mauvoisin. The results of our economic study show that the expected changes in climate in Switzerland may lead to a new type of reservoir management in a few decades. Finally, we examined recent trends and changes of the water concession regime in the Canton of Valais on the basis of available documents and a participatory procedure.

This experiment allowed to identify three types of strategies and debates which are related to (1) the assignment of responsibility, (2) the delimitation of uncertainties, (3) and the synchronisation of time horizons.

Relevance for research and practice

Climate and glacier changes, electricity market liberalization and the end of the water concessions cause new challenges. Important results could be provided to public and private decision makers at a national, cantonal and local level. They are indispensable in order to define consistent policies and strategies of water management for the medium and long term. Climate and glacier changes will considerably affect water inflows into reservoirs, as well as power and revenue generation. Moreover, since the 1990s, electricity markets have been opened to competition. To a large extent, the management of the water reservoirs will be more influenced by power price fluctuations than by runoff changes. To summarize, four lessons are highlighted:

  1. The results related to glacier and runoff changes are more optimistic than previously expected. However, stakeholders are still concerned by the impact of glacier retreat, not only on the availability of water, but also on sediments and risk management, which could impact both the production of electricity and the protection of population.
  2. Glacier retreat doesn’t only impact hydropower production. The question of glacier retreat raises the major issue of water availability and storage for all sorts of uses. Therefore they should be taken seriously and be further investigated: agriculture, flood protection, artificial snow production. It recalls that water is a resource which never supplies only one kind of activity.
  3. The better knowledge of the challenges of water availability is a key because it allows a better planning of the future prioritisation of uses. This project contributed to this knowledge, but more research should be done. Water resources should be considered in an integrated manner at the cantonal or national scale.
  4. Hydropower will be affected by climate change. However, runoff is only slightly influenced by greenhouse gas emissions scenarios. The main uncertainty lies in the wholesale electricity price. Therefore price uncertainty is a key variable for energy policy. New water concessions should consider this aspect and provide adequate flexibility to the potential investors.

Out of a scientific perspective, the physical modelling component of the project generated a key result: The type of model of glacier response and the modelling assumptions made in representing glacier processes can considerably affect future projections of glacier runoff. This calls for continuous improvements of the science behind these projections. It seems equally important that the physical constraints are strongly coupled with economic analysis in an interdisciplinary approach.

Original title

Future glacier evolution and consequences for the hydrology (FUGE)

Project leaders

  • Prof. Dr. Martin Funk, Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie (VAW), ETH Zürich
  • Dr. Andreas Bauder, Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie (VAW), ETH Zürich
  • Prof. Dr. Paolo Burlando, Institut für Umweltingenieurwissenschaften, ETH Zürich
  • Dr. Martin Lüthi, Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie (VAW), ETH Zürich
  • Dr. Francesca Pellicciotti, Institut für Umweltingenieurwissenschaften, ETH Zürich
  • Dr. Franco Romerio, Institut des sciences de l’environnement, Université de Genève

Further information on this content


Prof. Dr. Martin Funk Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie (VAW)
ETH Zürich
Gloriastrasse 37/39 8092 Zürich +41 44 632 41 32

Products of the project